JalView Tutorial - Jalview ? | jalview

JalView Tutorial - Jalview ? | jalview
Jalview 2.10.1
Manual and Introductory Tutorial
David Martin, James Procter
Andrew Waterhouse, Saif Shehata, Nancy Giang
Mungo Carstairs, Charles Ofoegbu, Kira Mourão
Suzanne Duce and Geoff Barton
School of Life Sciences, University of Dundee
Dundee, Scotland DD1 5EH, UK
Manual Version 1.9
20th February 2017
1 Basics
1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.1.1 Jalview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.1.2 Jalview’s Capabilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.1.3 About this Tutorial . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2 Launching the Jalview Desktop Application . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2.1 Getting Help . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.3 Navigation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.3.1 Navigation in Normal Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.3.2 Navigation in Cursor Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.3.3 The Find Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.4 Loading Sequences and Alignments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.4.1 Drag and Drop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.4.2 From a File . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.4.3 From a URL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.4.4 Cut and Paste . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.4.5 From a Public Database . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.4.6 Memory Limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
1.5 Saving Sequences and Alignments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.5.1 Saving Alignments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.5.2 Jalview Projects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2 Selecting and Editing Sequences
2.1 Selecting Parts of an Alignment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.1.1 Selecting Arbitrary Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.1.2 Selecting Columns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.1.3 Selecting Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.1.4 Making Selections in Cursor Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.1.5 Inverting the Current Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.2 Creating Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.3 Exporting the Current Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.4 Reordering an Alignment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.5 Hiding Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.5.1 Representing a Group with a Single Sequence . . . . . . . . . . . . . . . . . . . . . 22
2.6 Introducing and Removing Gaps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.6.1 Undoing Edits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.6.2 Locked Editing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.6.3 Introducing Gaps in a Single Sequence . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.6.4 Introducing Gaps in all Sequences of a Group . . . . . . . . . . . . . . . . . . . . . 24
2.6.5 Sliding Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.6.6 Editing in Cursor mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
3 Colouring Sequences and Figure Generation
3.1 Colouring Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.1.1 Colouring the Whole Alignment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.1.2 Colouring a Group or Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.1.3 Shading by Conservation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.1.4 Thresholding by Percentage Identity . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.1.5 Colouring by Annotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.1.6 Colour Schemes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.2 Formatting and Graphics Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.2.1 Multiple Alignment Views . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
3.2.2 Alignment Layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
3.2.3 Annotation Ordering and Display . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
3.2.4 Graphical Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
4 Annotation and Features
4.1 Conservation, Quality and Consensus Annotation . . . . . . . . . . . . . . . . . . . . . . . 39
4.1.1 Creating User Defined Annotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
4.1.2 Automated Annotation of Alignments and Groups . . . . . . . . . . . . . . . . . . . 41
4.2 Importing Features from Databases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
4.2.1 Sequence Database Reference Retrieval . . . . . . . . . . . . . . . . . . . . . . . . . 43
4.2.2 Colouring Features by Score or Description Text . . . . . . . . . . . . . . . . . . . . 44
4.2.3 Using Features to Re-order the Alignment . . . . . . . . . . . . . . . . . . . . . . . 45
4.2.4 Creating Sequence Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.2.5 Customising Feature Display . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
4.2.6 Sequence Feature File Formats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
5 Multiple Sequence Alignment
5.1 Performing a multiple sequence alignment . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
5.1.1 Realignment to add sequences to an existing alignment . . . . . . . . . . . . . . . 50
5.1.2 Alignments of Sequences that include Hidden Regions . . . . . . . . . . . . . . . . 50
5.1.3 Alignment Service Limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
5.2 Customising the Parameters used for Alignment . . . . . . . . . . . . . . . . . . . . . . . . 51
5.2.1 Getting Help on the Parameters for a Service . . . . . . . . . . . . . . . . . . . . . 52
5.2.2 Alignment Presets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
5.2.3 User Defined Presets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
5.3 Protein Alignment Conservation Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
5.3.1 Enabling and Disabling AACon Calculations . . . . . . . . . . . . . . . . . . . . . . 53
5.3.2 Configuring which AACon Calculations are Performed . . . . . . . . . . . . . . . . 54
5.3.3 Changing the Server used for AACon Calculations . . . . . . . . . . . . . . . . . . 54
6 Analysis of Alignments
6.1 PCA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
6.2 Trees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
6.2.1 Tree Based Conservation Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
6.2.2 Redundancy Removal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
6.2.3 Subdividing the Alignment According to Specific Mutations . . . . . . . . . . . . . 61
6.3 Pairwise Alignments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
7 Working with 3D structures
7.1 Molecular graphics systems supported by Jalview . . . . . . . . . . . . . . . . . . . . . . . 63
7.1.1 Configuring the default structure viewer . . . . . . . . . . . . . . . . . . . . . . . . 63
7.2 Automatic Association of PDB Structures with Sequences . . . . . . . . . . . . . . . . . . 64
7.2.1 Drag-and-Drop Association of PDB Files with Sequences by Filename Match . . 64
7.3 Viewing Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
7.3.1 Customising Structure Display . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
7.3.2 Superimposing Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
7.3.3 Colouring Structure Data Associated with Multiple Alignments and Views . . . 69
8 Protein sequence analysis and structure prediction
8.1 Protein Secondary Structure Prediction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
8.1.1 Hidden Columns and JPred Predictions . . . . . . . . . . . . . . . . . . . . . . . . . 74
8.2 Protein Disorder Prediction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
8.2.1 Disorder Prediction Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
8.2.2 Navigating Large Sets of Disorder Predictions . . . . . . . . . . . . . . . . . . . . . 76
8.2.3 Disorder Predictors provided by JABAWS 2.0 . . . . . . . . . . . . . . . . . . . . . 77
9 DNA and RNA Sequences
9.1 Working with DNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
9.1.1 Alignment and Colouring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
9.1.2 Translate cDNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
9.1.3 Linked DNA and Protein Views . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
9.1.4 Coding Regions from ENA Records . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
9.2 Working with RNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
9.2.1 Performing RNA Secondary Structure Predictions . . . . . . . . . . . . . . . . . . 85
10 Webservices
10.0.2 One-Way Web Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
10.0.3 Remote Analysis Web Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
10.0.4 JABA Web Services for Sequence Alignment and Analysis . . . . . . . . . . . . . . 88
10.0.5 Changing the Web Services Menu Layout . . . . . . . . . . . . . . . . . . . . . . . . 88
10.0.6 Running your own JABA Server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
Chapter 1
1.1 Introduction
1.1.1 Jalview
Jalview is a multiple sequence alignment viewer, editor and analysis tool. Jalview is designed to be
platform independent (running on Mac, MS Windows, Linux and any other platforms that support
Java). Jalview is capable of editing and analysing large alignments (thousands of sequences) with
minimal degradation in performance, and able to show multiple integrated views of the alignment
and other data. Jalview can read and write many common sequence formats including FASTA,
Clustal, MSF(GCG) and PIR.
There are two types of Jalview program. The Jalview Desktop is a standalone application that
provides powerful editing, visualization, annotation and analysis capabilities. The JalviewLite
applet has the same core visualization, editing and analysis capabilities as the desktop, without
the desktop’s webservice and figure generation capabilities. It is designed to be embedded in a web
page,1 and includes a javascript API to allow customisable display of alignments for web sites such
as Pfam.2
The Jalview Desktop in this version provides access to protein and nucleic acid sequence, alignment
and structure databases, and includes the Jmol3 viewer for molecular structures, and the VARNA4
program for the visualization of RNA secondary structure. It also provides a graphical user interface
for the multiple sequence alignment, conservation analysis and protein disorder prediction methods
provided as Java Bioinformatics Analysis Web Services (JABAWS). JABAWS5 is a system for running bioinformatics programs that you can download and run on your own machine or cluster, or
install on compute clouds.
1 A demonstration version of Jalview (Jalview Micro Edition) also runs on a mobile phone but the functionality is limited
to sequence colouring.
2 http://pfam.xfam.org
3 Provided under the LGPL licence at http://www.jmol.org
4 Provided under GPL licence at http://varna.lri.fr
5 released under GPL at
Jalview’s Capabilities
Figure 1.1 gives an overview of the main features of the Jalview desktop application. Its primary
function is the editing and visualization of sequence alignments, and their interactive analysis. Tree
building, principal components analysis, physico-chemical property conservation and sequence consensus analyses are built into the program. Web services enable Jalview to access online alignment
and secondary structure prediction programs, as well as to retrieve protein and nucleic acid sequences, alignments, protein structures and sequence annotation. Sequences, alignments, trees,
structures, features and alignment annotation may also be exchanged with the local filesystem.
Multiple visualizations of an alignment may be worked on simultaneously, and the user interface
provides a comprehensive set of controls for colouring and layout. Alignment views are dynamically
linked with Jmol and UCSF Chimera6 structure displays, a tree viewer and spatial cluster display,
facilitating interactive exploration of the alignment’s structure. The application provides its own
Jalview project file format in order to store the current state of an alignment and analysis windows.
Jalview also provides WYSIWIG7 style figure generation capabilities for the preparation of alignments for publication.
‘Standard’ Formats
 
 
 
Jalview Features
Jalview Annotation
Figure 1.1: Capabilities of the Jalview Desktop. The Jalview Desktop Application provides
a stable environment for the creation, editing and analysis of alignments and the
generation of figures.
6 UCSF Chimera needs to be installed separately. It is available free for academic use from
7 WYSIWIG: What You See Is What You Get.
Jalview History
Jalview was initially developed in 1996 by Michele Clamp, James Cuff, Steve Searle and Geoff Barton
at the University of Oxford and then the European Bioinformatics Institute. Development of Jalview
2 was made possible with eScience funding from the BBSRC8 in 2004, enabling Andrew Waterhouse
and Jim Procter to re-engineer the original program to introduce contemporary developments in
bioinformatics and take advantage of the latest web and Java technology. Jalview’s development
has been supported from 2009 onwards by BBSRC funding, and since 2014 by a Wellcome Trust
Biomedical Resource grant9 . In 2010, 2011, and 2012, Jalview benefitted from the Google Summer
of Code, when Lauren Lui and Jan Engelhardt introduced new features for handling RNA alignments
and secondary structure annotation, in collaboration with Yann Ponty.10
Citing Jalview
If you use Jalview in your work you should cite:
"Jalview Version 2 - a multiple sequence alignment editor and analysis workbench"
Waterhouse, A.M., Procter, J.B., Martin, D.M.A, Clamp, M. and Barton, G. J. (2009)
Bioinformatics doi: 10.1093/bioinformatics/btp033
This paper supersedes the original Jalview publication:
"The Jalview Java alignment editor"
Michele Clamp, James Cuff, Stephen M. Searle and Geoffrey J. Barton (2004)
Bioinformatics 20 426-427.
About this Tutorial
This tutorial is written in a manual format with short exercises where appropriate, typically at
the end of each section. The first few sections concerns the basic operation of Jalview and should
be sufficient for those who want to launch Jalview (Section 1.2), open an alignment (Section 1.4),
perform basic editing (Section 2), colouring (Section 3.1), and produce publication and presentation
quality graphical output (Section 3.2).
The remaining sections of the manual cover the visualization and analysis techniques available
in Jalview. These include working with the embedded Jmol molecular structure viewer (or UCSF
Chimera), building and viewing trees and Principal Components Analysis (PCA) plots, and using
trees for sequence conservation analysis. An overview of the Jalview Desktop’s webservices is given
in Section 10, and the alignment and secondary structure prediction services are described in detail
in Sections 5 and 8.1 respectively. Section 4 details the creation and visualization of sequence and
alignment annotation. Section 9.1 discusses specific features of use when working with nucleic acid
sequences, such as translation and linking to protein coding regions, and the display and analysis of
RNA secondary structure.
8 Biotechnology and Biological Sciences Research Council grant “VAMSAS: Visualization and Analysis of Molecules,
Sequence Alignments and Structures", a joint project to enable interoperability between Jalview, TOPALi and AstexViewer.
9 Wellcome grant number 101651/Z/13/Z
10 http://www.lix.polytechnique.fr/~ponty/
Typographic Conventions
Keystrokes using the special non-symbol keys are represented in the tutorial by enclosing the pressed
keys with square brackets (e.g. [RETURN] or [CTRL]).
Keystroke combinations are denoted with a ‘-’ symbol (e.g. [CTRL]-C means press [CTRL] and the ‘C’
key simultaneously).
Menu options are given as a path from the menu that contains them - for example File ⇒ Input
Alignment ⇒ From URL means to select the ‘From URL’ option from the ‘Input Alignment’ submenu
of a window’s ‘File’ dropdown menu.
Launching the Jalview Desktop Application
Figure 1.2: Download page on the Jalview web site at www.jalview.org.
This tutorial is based on the Jalview Desktop application. Much of the information will also be
useful for users of the JalviewLite applet, which has the same core editing, analysis and visualization
capabilities (see the JalviewLite Applet Examples page for examples). The Jalview Desktop, however,
is much more powerful, and includes additional support for interaction with external web services,
and production of publication quality graphics.
The Jalview Desktop can be run in two ways; as an application launched from the web via Java
webstart, or as an application loaded onto your hard drive. The webstart version is launched
from the pink ‘Launch Jalview Desktop button’ at the top right hand side of pages of the website
(www.jalview.org). To download the locally installable version, follow the links on the download page
(www.jalview.org/download) (Figure 1.2). These links will launch the latest stable release of Jalview.
When the application is launched with webstart, two dialogs may appear before the application
starts. If your browser is not set up to handle webstart, then clicking the launch link may download
a file that needs to be opened manually, or prompt you to select the program to handle the webstart
file. If that is the case, then you will need to locate the javaws program on your system11 . Once java
webstart has been launched, you may also be prompted to accept a security certificate signed by the
Barton Group.12 You can always trust us, so click trust or accept as appropriate. The splash screen
(Figure 1.3) gives information about the version and build date that you are running, information
about later versions (if available), and the paper to cite in your publications. This information is also
available on the Jalview web site at http://www.jalview.org.
Figure 1.3: Jalview splash screen.
When Jalview starts it will automatically load an example alignment from the Jalview site. This
behaviour can be switched off in the Jalview Desktop preferences dialog by unchecking the open file
option. This alignment will look like the one in Figure 1.4 (taken from Jalview version 2.10.1).
Jalview News RSS Feed
Announcements are made available to users of the Jalview Desktop via the Jalview Newsreader.
This window will open automatically when new news is available, and can also be accessed via the
Desktop’s Tools ⇒ Show Jalview News menu entry.
11 The file that is downloaded will have a type of application/x-java-jnlp-file or .jnlp. The javaws program that can
run this file is usually found in the bin directory of your Java installation
12 On some systems, the certificate may be signed by ’UNKNOWN’. In this case, clicking through the dialogs to look at
the detailed information about the certificate should reveal it to be a Barton group certificate.
Figure 1.4: Default startup for Jalview.
Figure 1.5: The Jalview News Reader. The newsreader opens automatically when new articles are available from the Jalview Desktop’s news channel.
Exercise 1: Launching Jalview from the Jalview Website
1.a. Open the Jalview web site (www.jalview.org) in your web browser. Launch Jalview
by clicking on the pink ‘Launch Jalview’ Desktop button in the top right hand corner.
This will download and open a jalview.jnlp webstart file.
1.b. Dialog boxes will open and ask if you want to open the jalview.jnlp file as the file is
an application downloaded from the Internet, click Open. (Note you may be asked to
update Java, if you agree then it will automatically update the Java software). As
Jalview opens, four demo Jalview windows automatically load.
1.c. If you are having trouble, it may help changing the browser you are using, as the
browsers and its version may affect this process.
1.d. To deactivate the opening of the 4 demo sequences during the launch, go to the Tools
⇒ Preferences... menu on the desktop. A ‘Preference’ dialog box opens, untick the box
adjacent to the ‘Open file’ entry in the ‘Visual’ preferences tab. Click OK to save the
1.e. Launch another Jalview workbench from the web site by clicking on the pink Launch
button. The example alignment should not be loaded as Jalview starts up.
1.f. To reload the original demo file select the File⇒ From URL entry in the Desktop
menu. Click on the URL history button (a downward arrow on the right hand side of
the dialog box) to view the files, select exampleFile_2_7.jar, then click OK.
Note: Should you want to load your own sequence during the launch process, then go to
the Tools ⇒ Preferences... menu on the desktop. The tick the ‘Open file’ entry of ‘Visual’
preferences tab, type in the URL of the sequence you want to load.
As the jalview.jnlp file launches Jalview on your desktop, you may want to move this from
the downloads folder to another folder. Opening from the jnlp file will allow Jalview to be
launched offline.
See the video at: http://www.jalview.org/Help/Getting-Started.
1.2.1 Getting Help
Built in Documentation
Jalview has comprehensive on-line help documentation. Select Help ⇒ Documentation from the
main window menu and a new window will open (Figure 1.6). The appropriate topic can then be
selected from the navigation panel on the left hand side. To search for a specific topic, click the
‘search’ tab and enter keywords in the box which appears.
Email Lists
The Jalview Discussion list [email protected] provides a forum for Jalview users and
developers to raise problems and exchange ideas - any problems, bugs, and requests for help should
be raised here. The [email protected] list can also be subscribed to if you wish to be
kept informed of new releases and developments.
Archives and mailing list subscription details can be found in the Jalview web site’s community
Figure 1.6: Accessing the built in Jalview documentation.
The major features of the Jalview Desktop are illustrated in Figure 1.7. The alignment window is
the primary window for editing and visualization, and can contain several independent views of the
alignment being worked with. The other windows (Trees, Structures, PCA plots, etc) are linked to a
specific alignment view. Each area of the alignment window has a separate context menu accessed
by clicking the right mouse button.
Jalview has two navigation and editing modes: normal mode, where editing and navigation is
performed using the mouse, and cursor mode where editing and navigation are performed using
the keyboard. The F2 key is used to switch between these two modes.
Note: On MacBooks and other laptops with compact keyboards, you may need to press the function
key [Fn] when pressing any of the numbered function keys. So to toggle between keyboard and
normal mode, press [Fn]-[F2].
Navigation in Normal Mode
Jalview always starts up in Normal mode, where the mouse is used to interact with the displayed
alignment view. You can move about the alignment by clicking and dragging the ruler scroll bar to
move horizontally, or by clicking and dragging the alignment scroll bar to the right of the alignment
to move vertically. If all the rows or columns in the alignment are displayed, the scroll bars will not
be visible.
Each alignment view shown in the alignment window presents a window onto the visible regions of
Desktop Window
Alignment Window
Tree Window
Alignment View Tabs
Sequence ID Panel
Alignment Ruler
Sequence Alignment
Alignment Scrollbar
Annotation Label Panel
Structure Window
Alignment Annotation
Status bar
Ruler Scrollbar
Figure 1.7: The anatomy of Jalview. The major features of the Jalview Desktop Application
are labeled.
the alignment. This means that with anything more than a few residues or sequences, alignments
can become difficult to visualize on the screen because only a small area can be shown at a time. It
can help, especially when examining a large alignment, to have an overview of the whole alignment.
Select View ⇒ Overview Window from the Alignment window menu bar (Figure 1.813 ).
The red box in the overview window shows the current view in the alignment window. A percent
identity histogram is plotted below the alignment overview. Shaded parts indicate rows and columns
of the alignment that are hidden (in this case, a single row at the bottom of the alignment - see
Section 2.5). You can navigate around the alignment by dragging the red box.
Alignment and analysis windows are closed by
clicking on the usual ‘close’ icon (indicated by
arrows on Mac OS X). If you want to close all
the alignments and analysis windows at once,
then use the Window ⇒ Close All option from
the Jalview desktop.
Warning: Make sure you have saved your
work because this cannot be undone!
13 the menu shown in this figure is from Jalview 2.2, later versions have more options.
Figure 1.8: Alignment Overview Window. The overview window for a view is opened from
the View menu.
Navigation in Cursor Mode
Cursor mode navigation enables the user to quickly and precisely navigate,
select and edit parts of an alignment. On pressing F2 to enter cursor mode
the position of the cursor is indicated by a black background and white text.
The cursor can be placed using the mouse or moved by pressing the arrow
keys (↑, ↓, ←, →).
Rapid movement to specific positions is accomplished as listed below:
◦ Jump to Sequence n: Type a number n then press [S] to move to sequence (row) n.
◦ Jump to Column n: Type a number n then press [C] to move to column n in the alignment.
◦ Jump to Residue n: Type a number n then press [P] to move to residue number n in the
current sequence.
◦ Jump to column m row n: Type the column number m, a comma, the row number n and
press [RETURN].
The Find Dialog Box
A further option for navigation is to use the Select ⇒ Find. . . function. This opens a dialog box into
which can be entered regular expressions for searching sequences and sequence IDs, or sequence
numbers. Hitting the [Find next] button will highlight the first (or next) occurrence of that pattern
in the sequence ID panel or the alignment, and will adjust the view in order to display the highlighted
region. The Jalview Help provides comprehensive documentation for this function, and a quick guide
to the regular expressions that can be used with it.
Exercise 2: Navigation
Jalview has two navigation and editing modes: normal mode (where editing and navigation
are via the mouse) and the cursor mode (where editing and navigation are via the keyboard).
The F2 key is used to switch between these two modes. With a Mac, the key combination
Fn key and F2 is needed, as button is often assigned to screen brightness. Jalview always
starts up in normal mode.
2.a. Load an example alignment from its URL (http://www.jalview.org/examples/
exampleFile_2_7.jar) via the Desktop using File ⇒ Input Alignment ⇒ From URL
dialog box. (The URL should be stored in its history and clicking on the down arrow
on the dialog box is an easy way to access it.)
2.b. Scroll around the alignment using the alignment (vertical) and ruler (horizontal)
scroll bars.
2.c. Find the Overview Window, Views ⇒ Overview Window and open it. Move around
the alignment by clicking and dragging the red box in the overview window.
2.d. Return to the alignment window, look at the status bar (lower left hand corner of the
alignment window) as you move the mouse over the alignment. It indicates information about the sequence and residue under the cursor.
2.e. Press [F2] key, or [Fn]-[F2] on Mac, to enter Cursor mode. Use the direction keys to
move the cursor around the alignment.
2.f. Move to sequence 7 by pressing 7 S. Move to column 18 by pressing 1 8 C. Move to
residue 18 by pressing 1 8 P. Note that these can be two different positions if gaps
are inserted into the sequence. Move to sequence 5, column 13 by typing 1 3 , 5
Note: To view Jalview’s comprehensive on-line help documentations select Help in desktop
menu, clicking on Documentation will open a Documentation window. Select topic from the
navigation panel on the left hand side or use the Search tab to select specific key words.
See the video at: http://www.jalview.org/Help/Getting-Started.
Loading Sequences and Alignments
Drag and Drop
In most operating systems you can just drag a file icon from a file browser window and drop it on an
open Jalview application window. The file will then be opened as a new alignment window. Drag and
drop also works when loading data from a URL - simply drag the link or url from the address panel
of your browser on to an alignment or the Jalview desktop background and Jalview will load data
from the URL directly.
1.4.2 From a File
Jalview can read sequence alignments from a sequence alignment file. This is a text file, not a word
processor document. For entering sequences from a wordprocessor document see Cut and Paste
(Section 1.4.4) below. Select File ⇒ Input Alignment ⇒ From File from the main menu (Figure 1.9).
You will then get a file selection window where you can choose the file to open. Remember to select
the appropriate file type. Jalview can automatically identify some sequence file formats.
Figure 1.9: Opening an alignment from a file saved on disk.
From a URL
Jalview can read sequence alignments directly from a URL. Please note that the files must be in a
sequence alignment format - an HTML alignment or graphics file cannot be read by Jalview. Select
File ⇒ Input Alignment ⇒ From URL from the main menu and a window will appear asking you to
enter the URL (Figure 1.10). Jalview will attempt to automatically discover the file format.
Figure 1.10: Opening an alignment from a URL.
Cut and Paste
Documents such as those produced by Microsoft Word cannot be readily understood by Jalview. The
way to read sequences from these documents is to select the data from the document and copy it to
the clipboard. There are two ways to do this. One is to right-click on the desktop background, and
select the ‘Paste to new window’ option in the menu that appears. The other is to select File ⇒ Input
Alignment ⇒ From Textbox from the main menu, paste the sequences into the text window that will
appear, and select New Window (Figure 1.11). In both cases, presuming that they are in the right
format, Jalview will happily read them into a new alignment window.
From a Public Database
Jalview can retrieve sequences and sequence alignments from the public databases housed at the
European Bioinformatics Institute, including Uniprot, Pfam, Rfam and the PDB. Jalview’s sequence
fetching capabilities allow you to avoid having to manually locate and save sequences from a web
Figure 1.11: Opening an alignment from pasted text.
page before loading them into Jalview. It also allows Jalview to gather additional metadata provided
by the source, such as annotation and database cross-references. Select File ⇒ Fetch Sequence(s) . . .
from the main menu and a window will appear (Figure 1.12). Pressing the database selection button
in the dialog box opens a new window showing all the database sources Jalview can access (grouped
by the type of database). Once you’ve selected the appropriate database, hit OK close the database
selection window, and then enter one or several database IDs or accession numbers separated by
a semicolon and press OK. Jalview will then attempt to retrieve them from the chosen database.
Example queries are provided for some databases to test that a source is operational, and can also
be used as a guide for the type of accession numbers understood by the source.
Figure 1.12: Retrieving sequences from a public database.
Memory Limits
Jalview is a Java program. One unfortunate implication of this is that Jalview cannot dynamically
request additional memory from the operating system. It is important, therefore, that you ensure
that you have allocated enough memory to work with your data. On most occasions, Jalview will warn
you when you have tried to load an alignment that is too big to fit in to memory (for instance, some
of the PFAM alignments are very large). You can find out how much memory is available to Jalview
with the desktop window’s ⇒ Tools ⇒ Show Memory Usage function, which enables the display of
the currently available memory at the bottom left hand side of the Desktop window’s background.
Should you need to increase the amount of memory available to Jalview, full instructions are given
in the built in documentation (opened by selecting Help ⇒ Documentation) and on the JVM memory
parameters page (http://www.jalview.org/jvmmemoryparams.html).
Exercise 3: Loading Sequences
3.a. Use Window ⇒ Close All from the Desktop window menu to close all windows.
3.b. Loading sequences from URL: Selecting File ⇒ Input Alignment ⇒ From URL
from the Desktop and enter http://www.jalview.org/tutorial/alignment.fa in the box.
Click OK to load the alignment.
3.c. Loading sequences from a file: Close all windows using the Window ⇒ Close All
menu option from the Desktop. Then type the same URL (http://www.jalview.
org/tutorial/alignment.fa) into your web browser and save the file to your desktop. Open the file you have just saved in Jalview by selecting File ⇒ Input Alignment
⇒ From File from the desktop menu and selecting this file. Click OK to load.
3.d. Loading sequences by ‘Drag and Drop’ / ‘Cut and Paste’:
(i) Select Desktop ⇒ Window ⇒ Close All. Then drag the alignment.fa file from the
desktop and drop it onto the Jalview window, the alignment should open.
(ii) Test the differences between (a) dragging the sequence onto the Jalview desktop
and (b) dragging the sequence onto an existing alignment window.
(iii) Open http://www.jalview.org/tutorial/alignment.fa in a web browser.
Drag the URL directly from browser onto Jalview desktop. (If the URL is downloaded, then locate the file in your download directory and open it in a text editor.)
3.e. The text editor: (i) Open the alignment.fa file using text editor. Copy the sequence
text from the file into the clipboard and paste it into the desktop background by rightclicking and selecting the Paste to New Window menu option.
(ii) In the text editor, copy the sequence text from alignment.fa into the clipboard
(usually via the browser’s Edit ⇒ Copy menu option).
(iii) In the Desktop menu, select File ⇒ Input Alignment ⇒ From Textbox. Paste the
clipboard into the large window using the Edit ⇒ Paste text box menu option. Click
New Window and the alignment will be loaded.
3.f. Loading sequences from Public Database: (i) Select File ⇒ Fetch Sequence(s)...
from the Desktop. The Select Database Retrieval Source dialog will open showing all
the database sources. Select the PFAM seed database and click OK.
(ii)Once a source has been selected, the New Sequence Fetcher window will open.
Enter the accession number PF03460 and click OK. An alignment of about 174 sequences should load.
3.g. These can be viewed using the Overview window accessible from View ⇒ Overview
Window. Several database IDs can be loaded by using semicolons to separate them.
See the video at: http://www.jalview.org/Help/Getting-Started
Saving Sequences and Alignments
Saving Alignments
Jalview allows alignments to be saved to file in a variety of formats so they can be restored at a
later date, passed to colleagues or analysed in other programs. From the alignment window menu
select File ⇒ Save As and a dialog box will appear (Figure 1.13). You can navigate to an appropriate
directory in which to save the alignment. Jalview will remember the last filename and format used
to save (or load) the alignment, enabling you to quickly save the file during or after editing by using
the File ⇒ Save entry. The File ⇒ Output To Textbox menu option allows the alignment to be copied
and pasted into other documents or web servers.
Jalview offers several different formats in which an alignment can be saved. Of these, only the
jalview project format (.jar or .jvp) will preserve the colours, groupings and other additional information in the alignment. The other formats produce text files containing just the sequences with
no visualization information, although some allow limited annotation and sequence features to be
stored (e.g. AMSA). Unfortunately, as far as we are aware only Jalview can read Jalview project files.
Figure 1.13: Saving alignments in Jalview to disk.
Jalview Projects
If you wish to save a complete Jalview session rather than
just a single alignment (e.g. because you have calculated trees
or multiple different alignments) then save your work as a
Jalview Project file (.jvp).14 From the main menu select File
⇒ Save Project and a file save dialog box will appear. Loading
a project will restore Jalview to exactly the view at which the
file was saved, complete with all alignments, trees, annotation
and displayed structures rendered appropriately.
14 Tip: Ensure that you have allocated plenty of memory to Jalview when working with large alignments in Jalview
projects. See Section 1.4.6 for how to do this.
Exercise 4: Saving Alignments
4.a. Launch Jalview afresh, or use Desktop ⇒ Window ⇒ Close all .
4.b. Load the ferredoxin alignment (PF03460) from PFAM (seed) (see Exercise 3).
4.c. Select File ⇒ Save As from the alignment window menu. Choose a location into which
to save the alignment and select your preferred format. All formats except Jalview
jvp can be viewed in a normal text editor (e.g. Notepad) or in a web browser. Enter a
file name and click Save.
4.d. Check this file by closing all windows and opening it with Jalview, or by browsing to
it with your web browser.
4.e. Repeat the previous steps saving the files in different file formats.
4.f. Select File ⇒ Output to Textbox ⇒ FASTA. Select and copy this alignment to the
clipboard using the textbox menu options Edit ⇒ Select All followed by Edit ⇒ Copy.
The alignment can then be pasted into any application of choice, e.g. a word processor
or web form.
4.g. Ensure at least one alignment window is active in Jalview. Open the overview window View ⇒ Overview Window and scroll red box to any part of the alignment. Select
File ⇒ Save Project from the main menu and save the project in a suitable folder.
4.h. Close all windows and then load the project via the File ⇒ Load Project menu option.
Observe how all the windows and positions are exactly as they were when they were
See the video at: http://www.jalview.org/Help/Getting-Started.
Chapter 2
Selecting and Editing Sequences
Jalview makes extensive use of selections - most of the commands available from its menus operate
on the currently selected region of the alignment, either to change their appearance or perform some
kind of analysis. This section illustrates how to make and use selections and groups.
Selecting Parts of an Alignment
Selections can be of arbitrary regions in an alignment, one or more complete columns, or one or more
complete sequences. A selected region can be copied and pasted as a new alignment using the Edit
⇒ Copy and Edit ⇒ Paste ⇒ To New Alignment in the alignment window menu options. To clear
(unselect) the selection press the [ESC] (escape) key.
Selecting Arbitrary Regions
To select part of an alignment, place the mouse at the top left corner of the region you wish to select.
Press and hold the mouse button and drag the mouse to the bottom right corner of the chosen region
then release the mouse button. A dashed red box appears around the selected region (Figure 2.1).
Figure 2.1: Selecting a region in an alignment.
Selecting Columns
To select the same residues in all sequences, click and drag along the alignment ruler. This selects
the entire column of the alignment. Ranges of positions from the alignment ruler can also be selected by clicking on the first position and then holding down the [SHIFT] key whilst clicking the
other end of the selection. Discontinuous regions can be selected by holding down [CTRL] and clicking on positions to add to the column selection. Note that each [CTRL]-Click (PC) or [CMD]-Click
(Mac) changes the current selected sequence region to that column, but adds to the column selection.
Selected columns are indicated by red highlighting in the ruler bar (Figure 2.2).
Figure 2.2: Selecting multiple columns in an alignment. The red highlighting on the alignment ruler marks the selected columns. Note that only the most recently selected
column has a dashed-box around it to indicate a region selection.
Selecting Sequences
Figure 2.3: Selecting multiple sequences in an alignment. Use [CTRL] or [SHIFT] to select
many sequences at once.
To select multiple complete sequences, click and drag the mouse down the sequence ID panel. The
same techniques as used for columns (above) can be used with [SHIFT]-Click for continuous and
[CTRL]-Click to select discontinuous ranges of sequences (Figure 2.3).
Making Selections in Cursor Mode
To define a selection in cursor mode (which is enabled by pressing [F2] when the alignment window
is selected), navigate to the top left corner of the proposed selection (using the mouse, the arrow
keys, or the keystroke commands described in Section 1.3.2). Pressing the [Q] key marks this as the
corner. A red outline appears around the cursor (Figure 2.4).
Navigate to the bottom right corner of the proposed selection and press the [M] key. This marks the
bottom right corner of the selection. The selection can then be treated in the same way as if it had
been created in normal mode.
Figure 2.4: Making a selection in cursor mode. Navigate to the top left corner (left), press
[Q], navigate to the bottom right corner and press [M] (right).
Inverting the Current Selection
The current sequence or column selection can be inverted, using Select ⇒ Invert Sequence/Column
Selection in the alignment window. Inverting the selection is useful when selecting large regions in
an alignment, simply select the region that is to be kept unselected, and then invert the selection.
This may also be useful when hiding large regions in an alignment (see Section 2.5 below). Instead
of selecting the columns and rows that are to be hidden, simply select the region that is to be kept
visible, invert the selection, then select View ⇒ Hide ⇒ Selected Region.
Creating Groups
Selections are lost as soon as a different region is selected. Groups can be created which are labeled
regions of the alignment. To create a group, first select the region which is to comprise the group.
Then click the right mouse button on the selection to bring up a context menu. Select Selection ⇒
Figure 2.5: Creating a new group from a selection.
Group ⇒ Edit name and description of current group1 then enter a name for the group in the dialog
box which appears.
By default the new group will have a box drawn around it. The appearance of the group can be
changed (see Section 3.1 below). This group will stay defined even when the selection is removed.
Exercise 5: Making Selections and Groups
5.a. Close windows.
Load the ferredoxin alignment (PF03460 from PFAM (seed)).
5.b. Selecting an arbitrary region. Choose a residue and place the mouse cursor on it
(residue information will show in alignment window status bar). Click and drag the
mouse to the bottom-right to create a selection. As you drag, a red box will ‘rubber
band’ out to show the extent of the selection. Release the mouse button and a red box
borders the selected region. Press [ESC] to clear this.
5.c. Select one sequence by clicking on the sequence ID panel. Note that the sequence
ID takes on a highlighted background and a red box appears around the selected
sequence. Hold down [SHIFT] and click another sequence ID a few positions above
or below. Note how the selection expands to include all the sequences between the
two positions on which you clicked. Hold down [CTRL] and then click on several sequences’ IDs - both selected and unselected. Note how unselected IDs are individually
added to the selection and previously selected IDs are individually deselected.
5.d. Select columns by clicking on the Alignment Ruler. Note that the selected column
is marked with a red box. Hold down [SHIFT] and click a column beyond. Note the
selection expands to include all the sequences between the two positions on which
you clicked.
5.e. Enter Cursor mode using [F2], or [Fn]-F2 for Macs. Navigate to column 59, row 1 by
pressing 5 9 , 1 [RETURN]. Press Q to mark this position. Navigate to column 65,
row 8 by pressing 6 5 , 8 [RETURN]. Press M to complete the selection. Note to clear
the selection press the [ESC] key.
5.f. To create a group from the selected the region, click the right mouse button when
mouse is on the selection, this opens a context menu in the alignment window.
Open the Selection ⇒ Edit New Group ⇒ Group Colour menu and select Percentage
Identity. This will turn the selected region into a group and colour it accordingly.
5.g. Hold down [CTRL] and use the mouse to select and deselect sequences in the alignment by clicking on their Sequence ID label. Note how the group expands to include
newly selected sequences, and the Percentage Identity colouring changes.
5.h. Another way to resize the group is by using the mouse to click and drag the righthand edge of the selected group.
5.i. The current selection can be exported and saved by right clicking the mouse when on
the text area to open the Sequence ID context menu. Follow the menus and pick an
output format (eg BLC) from the Selection ⇒ Output to Textbox . . . submenu.
5.j. In the Alignment output window that opens, try manually editing the alignment
before clicking the New Window button. This opens the edited alignment in a new
alignment window.
See the video at: http://www.jalview.org/training/Training-Videos.
1 In earlier versions of Jalview, this entry was variously ‘Group’, ‘Edit Group Name’, or ‘JGroupXXXXX’ (Where XXXXX
was some serial number).
Exporting the Current Selection
The current selection can be copied to the clipboard (in PFAM format). It can also be output to a
textbox using the output functions in the pop-up menu obtained by right clicking the current selection. The textbox enables quick manual editing of the alignment prior to importing it into a new
window (using the New Window button) or saving to a file with the File ⇒ Save As pulldown menu
option from the text box.
Reordering an Alignment
Sequence reordering is simple. Highlight the sequences to be moved then press the up or down arrow
keys as appropriate (Figure 2.6). If you wish to move a sequence up past several other sequences it
is often quicker to select the group past which you want to move it and then move the group rather
than the individual sequence.
Figure 2.6: Reordering the alignment. The selected sequence moves up one position on pressing the ↑ key.
Exercise 6: Reordering the Alignment
6.a. Close windows.
Load the ferredoxin alignment (PF03460 from PFAM (seed)).
6.b. Select one of the sequence in the sequence ID panel, use the up and down arrow keys
to alter the sequence’s position in the alignment. (Note that this will not work in
cursor mode)
6.c. To select and move multiple sequences, use hold [SHIFT], and select two sequences
separated by one or more un-selected sequences, repeat using the [CTRL] key. Note
how multiple sequences are grouped together when they are re-ordered using the up
and down arrow keys.
See the video at: http://www.jalview.org/training/Training-Videos.
Hiding Regions
It is sometimes convenient to exclude some sequences or residues in the alignment without actually
deleting them. Jalview allows sequences or alignment columns within a view to be hidden, and this
facility has been used to create the several different views in the example alignment file that is
loaded when Jalview is first started (See Figure 1.4).
To hide a set of sequences, select them and right-click the mouse on the selected sequence IDs to
bring up the context pop-up menu. Select Hide Sequences and the sequences will be concealed, with
a small blue triangle indicating their position (Figure 2.7). To unhide (reveal) the sequences, right
click on the triangle and select Reveal Sequences from the context menu.
Figure 2.7: Hiding Sequences Hidden sequences are represented by a small blue triangle in
the sequence ID panel.
A similar mechanism applies to columns (Figure 2.8). Selected columns (indicated by a red marker)
can be hidden and revealed in the same way via the context pop-up menu by right clicking on the
ruler bar. The hidden column selection is indicated by a small blue triangle in the ruler bar.
Figure 2.8: Hiding Columns Hidden columns are represented by a small blue triangle in the
ruler bar.
It is often easier to select the region that you intend to work with, rather than the regions that you
want to hide. In this case, select the required region and use the View ⇒ Hide ⇒ All but Selected
Region menu entry, or press [Shift]+[Ctrl]+H to hide the unselected region.
Representing a Group with a Single Sequence
Instead of hiding a group completely, it is sometimes useful to work with just one representative
sequence. The <Sequence ID> ⇒ Represent group with <Sequence ID> option from the sequence
ID pop-up menu enables this variant of the hidden groups function. The remaining representative
sequence can be visualized and manipulated like any other. However, any alignment edits that affect
the sequence will also affect the whole sequence group.
Exercise 7: Hiding and Revealing Regions
7.a. Close windows.
Load the ferredoxin alignment (PF03460 from PFAM (seed)).
7.b. Select a contiguous set of sequences by clicking and dragging on the sequence ID
panel. Right click on the selected sequence IDs to bring up the sequence ID context
menu, select Hide Sequences.
7.c. Right click on the blue triangle indicating hidden sequences and select Reveal Sequences in the panel. (If you have hidden all sequences then you will need to use the
alignment window menu option View ⇒ Show ⇒ All Sequences.)
7.d. Repeat the process but use a non-contiguous set of sequences. Note that when multiple regions are hidden you can select either Reveal Sequences to reveal the hidden
sequences that were clicked, or Reveal All.
7.e. Repeat the above using columns to hide and reveal columns instead of sequences.
7.f. Select a region of the alignment, and experiment with the Hide all but selected region
option in View ⇒ Hide ⇒ All but selected region.
7.g. Select some sequences, pick one to represent the rest by hovering the mouse over
this sequence. Bring up the Sequence ID context menu by right clicking and select
(Sequence ID name) ⇒ Represent group with (Sequence ID name ). To reveal these
hidden sequences, right click on the Sequence ID and in the context menu select
Reveal All.
See the video at: http://www.jalview.org/training/Training-Videos.
Figure 2.9: Introducing gaps in a single sequence. Gaps are introduced as the selected
sequence is dragged to the right while pressing and holding [SHIFT].
Figure 2.10: Introducing gaps in a group. Gaps are introduced as the selected group is
dragged to the right with [CTRL] pressed.
Introducing and Removing Gaps
The alignment view provides an interactive editing interface, allowing gaps to be inserted or deleted
to the left of any position in a sequence or sequence group. Alignment editing can only be performed
whilst in keyboard editing mode (entered by pressing [F2]) or by clicking and dragging residues with
the mouse when [SHIFT] or [CTRL] is held down (which differs from earlier versions of Jalview).
2.6.1 Undoing Edits
Alignment edits can be undone via the Edit ⇒ Undo Edit alignment window menu option, or CTRLZ. An edit, if undone, may be re-applied with Edit ⇒ Redo Edit, or CTRL-Y. Note, however, that the
Undo function only works for edits to the alignment or sequence ordering. Colouring of the alignment, showing and hiding of sequences or modification of annotation that only affect the alignment’s
display cannot be undone.
2.6.2 Locked Editing
The Jalview alignment editing model is different to that used in other alignment editors. Because
edits are restricted to the insertion and deletion of gaps to the left of a particular sequence position,
editing has the effect of shifting the rest of the sequence(s) being edited down or up-stream with respect to the rest of alignment. The Edit ⇒ Pad Gaps option can be enabled to eliminate ‘ragged edges’
at the end of the alignment, but does not avoid the ‘knock-on’ effect which is sometimes undesirable.
However, its effect can be limited by performing the edit within a selected region. In this case, gaps
will only be removed or inserted within the selected region. Edits are similarly constrained when
they occur adjacent to a hidden column.
2.6.3 Introducing Gaps in a Single Sequence
To introduce a gap, first select the sequence in the sequence ID panel and then place the cursor on
the residue to the immediate right of where the gap should appear. Hold down the SHIFT key and
the left mouse button, then drag the sequence to the right until the required number of gaps has
been inserted.
One common error is to forget to hold down [SHIFT]. This results in a selection which is one sequence
high and one residue long. Gaps cannot be inserted in such a selection. The selection can be cleared
and editing enabled by pressing the [ESC] key.
2.6.4 Introducing Gaps in all Sequences of a Group
To insert gaps in all sequences in a selection or group, select the required sequences in the sequence
ID panel and then place the mouse cursor on any residue in the selection or group to the immediate
right of the position in which a gap should appear. Hold down the CTRL key and the left mouse
button, then drag the sequences to the right until the required number of gaps has appeared.
Gaps can be removed by dragging the residue to the immediate right of the gap leftwards whilst
holding down [SHIFT] (for single sequences) or [CTRL] (for a group of sequences).
Exercise 8: Editing Alignments
You are going to manually reconstruct part of the example Jalview alignment available at
Mac Users: Please use the Apple or [CMD] key in place of [CTRL] for key combinations such
as [CTRL]-A.
Remember to use [CTRL]-Z to undo an edit, or the File ⇒ Reload function to revert the
alignment back to the original version if you want to start again.
8.a. Load the URL http://www.jalview.org/tutorial/unaligned.fa which contains part of the
ferredoxin alignment from PF03460.
8.b. Select the first 7 sequences, and press H to hide them (or right click on the sequence
IDs to open the sequence ID context menu, and select Hide Sequences).
8.c. Select FER3_RAPSA and FER_BRANA. Slide the sequences to the right so the initial
residue A lies at column 57 using the ⇒ key.
8.d. Select FER1_SPIOL, FER1_ARATH, FER2_ARATH, Q93Z60_ARATH and
Hint: you can do this by pressing [CTRL]-I to invert the sequence selection and then
deselect FER1_MAIZE), and use the ⇒ key to slide them to so they begin at column
5 of the alignment view.
8.e. Select all the visible sequences (those not hidden) in the block by pressing [CTRL]-A.
Insert a single gap in all selected sequences at column 38 of the alignment by holding
[CTRL] and clicking on the R at column 38 in the FER1_SPIOL, then drag one column
to right. Insert another gap at column 47 in all sequences in the same way.
8.f. Correct the ferredoxin domain alignment for FER1_SPIOL by inserting two additional gaps after the gap at column 47. First press [ESC] to clear the selection, then
hold [SHIFT] and click and drag on the G and move it two columns to the right.
8.g. Now complete the alignment of FER1_SPIOL with a locked edit by pressing [ESC]
and select columns 47 to 57 of the FER1_SPIOL row. Move the mouse onto the G at
column 50, hold [SHIFT] and drag the G in column 47 of FER1_SPIOL to the left by
one column to insert a gap at column 57.
8.h. In the next two steps you will complete the alignment of the last two sequences.
Select the last two sequences (FER1_MAIZE and O80429_MAIZE), then press
[SHIFT] and click and drag the initial methionine of O80429_MAIZE 5 columns to
the right so it lies at column 10.
Keep holding [SHIFT] and click and drag to insert another gap at the proline at
column 25 (25C in cursor mode). Remove the gap at column 44, and insert 4 gaps at
column 47 (after AAPM).
8.i. Hold [SHIFT] and drag the I at column 39 of FER1_MAIZE 2 columns to the right.
Remove the gap at FER1_MAIZE column 49 by [SHIFT]-click and drag left by one column. Press [ESC] to clear the selection, and then insert three gaps in FER1_MAIZE
at column 47 by holding [SHIFT] and click and drag the S in FER1_MAIZE to the
right by three columns. Finally, remove the gap in O80429_MAIZE at column 56
using [SHIFT]-drag to the left on 56C.
8.j. Use the Edit ⇒ Undo Edit and Edit ⇒ Redo Edit menu option, or their keyboard
shortcuts ([CTRL]-Z and [CTRL]-Y) to step backwards and replay the edits you have
2.6.5 Sliding Sequences
Pressing the [←] or [→] arrow keys when one or more sequences are selected will “slide” the entire
selected sequences to the left or right (respectively). Slides occur regardless of the region selection which, for example, allows you to easily reposition misaligned subfamilies within a larger alignment.
2.6.6 Editing in Cursor mode
Gaps can be easily inserted when in cursor mode (toggled with [F2]) by pressing [SPACE]. Gaps
will be inserted at the cursor, shifting the residue under the cursor to the right. To insert n gaps
type n and then press [SPACE]. To insert gaps into all sequences of a group, use [CTRL]-[SPACE] or
[SHIFT]-[SPACE] (both keys held down together).
Gaps can be removed in cursor mode by pressing [BACKSPACE]. First make sure you have everything unselected by pressing ESC. The gap under the cursor will be removed. To remove n gaps, type
n and then press [BACKSPACE]. Gaps will be deleted up to the number specified. To delete gaps
from all sequences of a group, press [CTRL]-[BACKSPACE] or [SHIFT]-[BACKSPACE] (both keys
held down together). Note that the deletion will only occur if the gaps are in the same columns in all
sequences in the selected group, and those columns are to the right of the selected residue.
Exercise 9: Keyboard Edits
This continues on from the previous exercise, and recreates the final part of the example
ferredoxin alignment from the unaligned sequences using Jalview’s keyboard editing mode.
Window users: Please only use [SHIFT]-[SPACE] in this exercise.
Mac users: [CTRL]-[SPACE] can also be used instead of [SHIFT]-[SPACE].
9.a. Load the sequence alignment at http://www.jalview.org/tutorial/unaligned.fa, or continue using the edited alignment. If you continue from the previous exercise, first
right click on the sequence ID panel and select Reveal All. Enter cursor mode by
pressing [F2].
9.b. Insert 58 gaps at the start of the sequence 1 (FER_CAPAA). Press 58 then [SPACE].
9.c. Go down one sequence and select rows 2-5 as a block. Click on the second sequence
ID (FER_CAPAN). Hold down shift and click on the fifth (FER1_PEA).
9.d. Insert 6 gaps at the start of this group. Go to column 1 row 2 by typing 1,2 then press
[RETURN]. Now insert 6 gaps in all the sequences. Type 6 then hold down [SHIFT]
and press [SPACE].
9.e. Now insert one gap at column 34 and another at 38. Insert 3 gaps at 47. Press 34C
then [SHIFT]-[SPACE]. Press 38C then [SHIFT]-[SPACE]. Press 47C then 3 [SHIFTSPACE] the first through fourth sequences are now aligned.
9.f. The fifth sequence (FER1_PEA) is poorly aligned. We will delete some gaps and add
some new ones. Press [ESC] to clear the selection. Navigate to the start of sequence
5 and delete 3 gaps. Press 1,5 [RETURN] then 3 [BACKSPACE] to delete three gaps.
Go to column 31 and delete the gap. Press 31C [BACKSPACE] .
9.g. Similarly delete the gap now at column 34, then insert two gaps at column 38. Press
34C [BACKSPACE] 38C 2 [SPACE]. Delete three gaps at column 44 and insert one
at column 47 by pressing 44C 3 [BACKSPACE] 47C [SPACE]. The top five sequences
are now aligned.
Chapter 3
Colouring Sequences and Figure
Colouring Sequences
Colouring sequences is a key aspect of alignment presentation. Jalview allows you to colour the
whole alignment, or just specific groups. Alignment and group colours are rendered below any other
colours, such as those arising from sequence features (these are described in Section 4). This means
that if you try to apply one of the colourschemes described in this section, and nothing appears to
happen, it may be that you have sequence feature annotation displayed, and you may have to disable
it using the View ⇒ Show Features option before you can see your colourscheme.
There are two main types of colouring styles: simple static residue colourschemes and dynamic
schemes which use conservation and consensus analysis to control colouring. Hybrid colouring
is also possible, where static residue schemes are modified using a dynamic scheme. The individual
schemes are described in Section 3.1.6 below.
Colouring the Whole Alignment
The alignment can be coloured via the Colour menu option in the alignment window. Selecting the colour scheme
causes all residues to be coloured. The menu is divided
into three sections. The first section gives options for the
behaviour of the menu options, the second lists static and
dynamic colourschemes available for selection. The last
gives options for making hybrid colourschemes using conservation shading or colourscheme thresholding.
Colouring a Group or Selection
Selections or groups can be coloured in two ways. The first is via the Alignment Window’s Colour
menu as stated above, after first ensuring that the Apply Colour To All Groups flag is not selected.
This must be turned off specifically as it is on by default. When unticked, selections from the Colours
menu will only change the colour for residues in the current selection, or the alignment view’s “background colourscheme” when no selection exists.
The second method is to select sequences and right click mouse to open pop-up menu and select
Selection ⇒ Edit New Group ⇒ Group Colour from context menu options (Figure 3.1). This only
changes the colour of the current selection or group.
Figure 3.1: Colouring a group via the context menu.
Shading by Conservation
For many colour schemes, the intensity of the colour in a column can be scaled by the degree of amino
acid property conservation. Selecting Colour ⇒ By Conservation enables this mode, and Modify
Conservation Threshold... brings up a selection box (the Conservation Colour Increment dialog box)
allowing the alignment colouring to be modified. Selecting a higher value limits colouring to more
highly conserved columns (Figure 3.2).
Figure 3.2: Conservation Shading The density of the ClustalX style residue colouring is controlled by the conservation threshold. The effect of 0% (left), 50% (center) and 100%
(right) thresholds are shown.
Thresholding by Percentage Identity
‘Thresholding’ is another hybrid colour model where a residue is only coloured if it is not excluded by
an applied threshold. Selecting Colour ⇒ Above Identity Threshold brings up a selection box with
a slider controlling the minimum percentage identity threshold to be applied. Selecting a higher
threshold (by sliding to the right) limits the colouring to columns with a higher percentage identity
(as shown by the Consensus histogram in the annotation panel).
Colouring by Annotation
Any of the quantitative annotations shown on
an alignment can be used to threshold or shade
the whole alignment.1
The Colour ⇒ By Annotation option opens a dialog which allows you to select which annotation to use, the minimum and maximum shading
colours or whether the original colouring should
be thresholded (the ‘Use original colours’ option).
Default settings for minimum and maximum
colours can be set in the Jalview Desktop’s preferences.
The per Sequence option in the Colour By Annotation dialog allows each sequence to be shaded
according to sequence associated annotation rows, such as protein disorder scores. This functionality
is described further in Section 8.2.
Colour Schemes
Full details on each colour scheme can be found in the Jalview on-line help. A brief description of
each one is provided below:
This is an emulation of the default colourscheme used
for alignments in ClustalX, a graphical interface for
the ClustalW multiple sequence alignment program.
Each residue in the alignment is assigned a colour if
the amino acid profile of the alignment at that position
meets some minimum criteria specific for the residue
1 Please remember to turn off Sequence Feature display to see the shading
Gaps are coloured white. If a residue matches the consensus sequence residue at that position it is coloured
dark blue. If it does not match the consensus residue
but the Blosum62 matrix gives a positive score, it is
coloured light blue.
Percentage Identity
The Percent Identity option colours the residues
(boxes and/or text) according to the percentage of the
residues in each column that agree with the consensus sequence. Only the residues that agree with the
consensus residue for each column are coloured.
The residues are coloured according to their physicochemical properties. The physicochemical groupings are Aliphatic/hydrophobic, Aromatic, Positive,
Negative, Hydrophillic, conformationally special, and
This colour scheme was devised by Willie Taylor and
an entertaining description of its origin can be found
in Protein Engineering, Vol 10 , 743-746 (1997).
Residues are coloured according to the hydrophobicity table of Kyte, J., and Doolittle, R.F., J. Mol. Biol.
1157, 105-132, 1982. The most hydrophobic residues
are coloured red and the most hydrophilic ones are
coloured blue.
Helix Propensity
The residues are coloured according to their ChouFasman2 helix propensity. The highest propensity is
magenta, the lowest is green.
Strand Propensity
The residues are coloured according to their ChouFasman2 Strand propensity. The highest propensity
is Yellow, the lowest is blue.
Turn Propensity
The residues are coloured according to their ChouFasman2 turn propensity. The highest propensity is
red, the lowest is cyan.
Buried Index
The residues are coloured according to their ChouFasman2 burial propensity. The highest propensity is
blue, the lowest is green.
Residues are coloured with four colours corresponding
to the four nucleotide bases. All non ACTG residues
are uncoloured. See Section 9.1 for further information about working with nucleic acid sequences and
2 Chou, PY and Fasman, GD. Annu Rev Biochem. 1978;47:251-76.
Purine Pyrimidine
Residues are coloured according to whether the corresponding nucleotide bases are purine (magenta) or
pyrimidine (cyan) based. All non ACTG residues are
uncoloured. For further information about working
with nucleic acid sequences and alignments, see Section 9.1.
RNA Helix Colouring
Columns are coloured according to their assigned RNA
helix as defined by a secondary structure annotation
line on the alignment. Colours for each helix are randomly assigned, and option only available when an
RNA secondary structure row is present on the alignment.
User Defined
This dialog allows the user to create any number of named colour schemes at will. Any residue may
be assigned any colour. The colour scheme can then be named. If you save the colour scheme, this
name will appear on the Colour menu (Figure 3.3).
Figure 3.3: Creation of a user defined colour scheme. Residue types are assigned colours
(left). The profile is saved (center) and can then be accessed via the Colour menu
Exercise 10: Colouring Alignments
Note: Before you begin this exercise, ensure that the Apply Colour To All Groups flag is not
selected in Colour menu in the alignment window.
10.a. Open a sequence alignment, for example the PFAM domain PF03460 in PFAM seed
database. Select the alignment menu option Colour ⇒ ClustalX and note the colour
change. Now try all the other colour schemes in the Colour menu. Note that some
colour schemes do not colour all residues.
10.b. Colour the alignment using Colour ⇒ Blosum62. Select a group of around 4 similar sequences. Use the context menu (right click on the group) option Selection ⇒
Edit New Group ⇒ Group Colour ⇒ Blosum62 to colour the selection. Notice how
some residues which were not coloured are now coloured. The calculations performed
for dynamic colouring schemes like Blosum62 are based on the selected group, not
the whole alignment (this also explains the colouring changes observed in exercise 5
during the group selection step).
10.c. Keeping the same selection as before, colour the complete alignment except the group
using Colour ⇒ Taylor. Select the menu option Colour ⇒ By Conservation. Slide
the selector in the Conservation Colour Increment dialog box from side to side and
observe the changes in the alignment colouring in the selection and in the complete
Note: Feature colours overlay residue colouring. The features colours can be toggled off by
going to View ⇒ Show Sequence Features.
See the video at: http://www.jalview.org/training/Training-Videos.
Exercise 11: User Defined Colour Schemes
11.a. Load a sequence alignment. Select the alignment menu option Colour ⇒ User Defined. A dialog window will open.
11.b. Click on an amino acid button, then select a colour for that amino acid. Repeat till all
amino acids are coloured to your liking.
11.c. Insert a name for the colourscheme in the appropriate field and click Save Scheme.
You will be prompted for a file name in which to save the colour scheme. The dialog
window can now be closed.
11.d. The new colour scheme appears in the list of colour schemes in the Colour menu and
can be selected in future Jalview sessions.
See the video at: http://www.jalview.org/training/Training-Videos.
Formatting and Graphics Output
Jalview is a WYSIWIG alignment editor. This means that for most kinds of graphics output, the
layout that is seen on screen will be the same as what is outputted in an exported graphics file. It is
therefore important to pick the right kind of display layout prior to generating figures.
Multiple Alignment Views
Jalview is able to create multiple independent visualizations of the same underlying alignment these are called Views. Because each view displays the same underlying data, any edits performed
in one view will update the alignment or annotation visible in all views.
Alignment views are created using the View ⇒ New View option of the alignment window or by Pressing [CTRL]-T. This
will create a new view with the same groups, alignment layout and display options as the current one. Pressing G will
gather together Views as named tabs on the alignment window, and pressing X will expand gathered Views so they can
be viewed simultaneously in their own separate windows. To
delete a group, press [CTRL]-W.
Alignment Layout
Jalview provides two screen layout modes, unwrapped (the default) where the alignment is in one
long line across the window, and wrapped, where the alignment is on multiple lines, each the width of
the window. Most layout options are controlled by the Format menu option in the alignment window,
and control the overall look of the alignment in the view (rather than just a selected region).
Wrapped Alignments
Wrapped alignments can be toggled on and off using the Format ⇒ Wrap menu option (Figure 3.4).
Note that the annotation lines are also wrapped. Wrapped alignments are great for publications and
presentations but are of limited use when working with large numbers of sequences.
If annotations are not all visible in wrapped mode, expand the alignment window to view them. Note
that alignment annotation (see Section 4) cannot be interactively created or edited in wrapped mode,
and selection of large regions is difficult.
The text appearance in a view can be modified via the Format
⇒ Font. . . alignment window menu. This setting applies for
all alignment and annotation text except for that displayed
in tool-tips. Additionally, font size and spacing can be adjusted rapidly by clicking the middle mouse button and dragging across the alignment window.
Figure 3.4: Wrapping the alignment.
Numbering and Label Justification
Options in the Format menu are provided to control the alignment view, and provide a range of
options to control the display of sequence and alignment numbering, the justification of sequence
IDs and annotation row column labels on the annotation rows shown below the alignment.
Alignment and Group Colouring and Appearance
The display of hidden row/column markers and gap characters can be turned off with Format ⇒ Hidden Markers and Format ⇒ Show Gaps, respectively. The Text and Colour Text option controls the
display of sequence text and the application of alignment and group colouring to it. Boxes controls
the display of the background area behind each residue that is coloured by the applied coloursheme.
Highlighting Nonconserved Symbols
The alignment layout and group sub-menu both contain an option to hide conserved symbols from the
alignment display (Format ⇒ Show nonconserved in the alignment window or Selection ⇒ Group ⇒
Show Nonconserved by right clicking on a group). This mode is useful when working with alignments
that exhibit a high degree of homology, because Jalview will only display gaps or sequence symbols
that differ from the consensus for each column, and render all others with a ‘.’.
Figure 3.5: Hiding Annotations Annotations can either be hidden from the View menu (left)
or individually from the context menu (right).
Annotation Ordering and Display
The annotation lines which appear below the sequence alignment are described in detail in Section
4. They can be hidden by toggling the View ⇒ Show Annotations menu option. Additionally, each
annotation line can be hidden and revealed in the same way as sequences via the pop-up context
menu on the annotation name panel (Figure 3.5). Annotations can be reordered by dragging the
annotation line label on the annotation label panel. Placing the mouse over the top annotation label
brings up a resize icon on the left. When this is displayed, Click-dragging up and down provides
more space in the alignment window for viewing the annotations, and less space for the sequence
Exercise 12: Alignment Layout
12.a. Start Jalview and open the URL http://www.jalview.org/examples/exampleFile.jar. Select Format ⇒ Wrap from the alignment window menu. Experiment with the various
options from the Format menu, for example adjust the ruler placement, sequence ID
format and so on.
12.b. Hide all the annotation rows by toggling Annotations ⇒ Show Annotations from the
alignment window menu. Reveal the annotations by selecting the same menu option.
12.c. Deselect Format ⇒ Wrap. Right click on the annotation row labels to bring up the
context menu, then select Hide This Row. Bring up the context menu again and select
Show All Hidden Rows to reveal them.
12.d. Annotations can be reordered by clicking and dragging the row to the desired position.
Click on the Consensus row and drag it upwards to just above Quality. The rows
should now be reordered. Features and annotations are covered in more detail in
Section 4.
12.e. Move the mouse to the top left hand corner of the annotation labels - a grey up/down
arrow symbol should appear - when this is shown, the height of the Annotation Area
can be changed by clicking and dragging this icon up or down.
Graphical Output
Jalview allows alignments figures to be exported in three different formats, each of which is suited to a particular purpose.
Image export is via the File ⇒ Export Image ⇒ . . . alignment
window menu option.
HTML is the format used by web pages.
Jalview outputs the alignment as an HTML
table with all the colours and fonts as seen.
Any additional annotation will also be embedded as sensitive areas on the page, such as
URL links for each sequence’s ID label. This
file can then be viewed directly with any web
browser. Each residue is placed in an individual table cell. Unwrapped alignments will produce a very wide page.
     
EPS is Encapsulated Postscript. It is the
    
format of choice for publications
posters as it gives the highest 
quality output
    
of any of the image types. It can be scaled to
    
any size, so will still look good on an A0 poster.
    
This format can be read by most
good presentation and graphics packages such
as Adobe
    
Illustrator or Inkscape.
Zoom Detail of
 
EPS image.
PNG is Portable Network Graphics. This output option produces an image that can be easily included in web pages and incorporated in
presentations using e.g. Powerpoint or Open
Office. It is a bitmap image so does not scale
and is unsuitable for use on posters, or in publications.
For submission of alignment figures to journals, please use EPS3 .
Zoom Detail of PNG image.
Exercise 13: Graphical Output
13.a. Load the example Jalview Jar file in Exercise 12. Customise it how you wish but
leave it unwrapped. Select File ⇒ Export Image ⇒ HTML from the alignment menu.
Save the file and open it in your favoured web browser.
13.b. Wrap the alignment and export the image to HTML again. Compare the two images.
(Note that the exported image matches the format displayed in the alignment window
but annotations are not exported).
13.c. Export the alignment using the File ⇒ Export Image ⇒ PNG menu option. Open the
file in an image viewer that allows zooming such as Paint or Photoshop (Windows),
or Preview (Mac OS X) and zoom in. Notice that the image is a bitmap and it becomes pixelated when zoomed. (Note that the annotation lines are included in
the image.)
13.d. Export the alignment using the File ⇒ Export Image ⇒ EPS menu option. Open
the file in a suitable program such as Photoshop, Illustrator, Inkscape, Ghostview,
Powerpoint (Windows), or Preview (Mac OS X). Zoom in and note that the image has
near-infinite resolution.
3 If the journal complains, insist.
Chapter 4
Annotation and Features
Annotations and features are additional information that is overlaid on the sequences and the alignment. Generally speaking, annotations reflect properties of the alignment as a whole, often associated with columns in the alignment. Features are often associated with specific residues in the
Annotations are shown below the alignment in the annotation panel, the properties are often based
on the alignment. Conversely, sequence features are properties of the individual sequences, so they
do not change with the alignment, but are shown mapped on to specific residues within the alignment.
Features and annotation can be interactively created, or retrieved from external data sources. Webservices like JPred (see 8.1 above) can be used to analyse a given sequence or alignment and generate
annotation for it.
Conservation, Quality and Consensus Annotation
Jalview automatically calculates several quantitative alignment annotations which are displayed as
histograms below the multiple sequence alignment columns. Conservation, quality and consensus
scores are examples of dynamic annotation, so as the alignment changes, they change along with
it. The scores can be used in the hybrid colouring options to shade the alignments. Mousing over a
conservation histogram reveals a tooltip with more information.
These annotations can be hidden and deleted via the context menu linked to the annotation row; but
they are only created on loading an alignment. If they are deleted then the alignment should be saved
and then reloaded to restore them. Jalview provides a toggle to autocalculate a consensus sequence
upon editing. This is normally selected by default, but can be turned off for large alignments via the
Calculate ⇒ Autocalculate Consensus menu option if the interface is too slow.
Conservation Annotation
Alignment conservation annotation is quantitative numerical index reflecting the conservation of
the physico-chemical properties for each column of the alignment. The calculation is based on AMAS
method of multiple sequence alignment analysis (Livingstone C.D. and Barton G.J. (1993) CABIOS
Vol. 9 No. 6 p745-756), with identities scoring highest, and amino acids with substitutions in the
same physico-chemical class have next highest score. The score for each column is shown below the
histogram. The conserved columns with a score of 11 are indicated by ’*’. Columns with a score of 10
have mutations but all properties are conserved are marked with a ’+’.
Consensus Annotation
Alignment consensus annotation reflects the percentage of the different residue per column. By default this calculation includes gaps in columns, gaps can be ignored via the Consensus label context
menu to the left of the consensus bar chart. The consensus histogram can be overlaid with a sequence logo that reflects the symbol distribution at each column of the alignment. Right click on the
Consensus annotation row and select the Show Logo option to display the Consensus profile for the
group or alignment. Sequence logos can be enabled by default for all new alignments via the Visual
tab in the Jalview desktop’s preferences dialog box.
Quality Annotation
Alignment quality annotation is an ad-hoc measure of the likelihood of observing the mutations (if
any) in a particular column of the alignment. The quality score is calculated for each column in an
alignment by summing, for all mutations, the ratio of the two BLOSUM 62 scores for a mutation pair
and each residue’s conserved BLOSUM62 score (which is higher). This value is normalised for each
column, and then plotted on a scale from 0 to 1.
Group Associated Annotation
Group associated consensus and conservation annotation rows reflect the sequence variation within
a particular group. Their calculation is enabled by selecting the Group Conservation or Group Consensus options in the Annotation ⇒ Autocalculated Annotation submenu of the alignment window.
Creating User Defined Annotation
To create a new annotation row, right click on the annotation label panel and select the Add New
Row menu option (Figure 4.1). A dialog box appears. Enter the label to use for this row and a new
row will appear.
To create a new annotation, first select all the positions to be annotated on the appropriate row.
Right-clicking on this selection brings up the context menu which allows the insertion of graphics for
secondary structure (Helix or Sheet), text Label and the colour in which to present the annotation
(Figure 4.2). On selecting Label a dialog box will appear, requesting the text to place at that position.
After the text is entered, the selection can be removed and the annotation becomes clearly visible1 .
Annotations can be coloured or deleted as desired.
Figure 4.1: Creating a new annotation row. Annotation rows can be reordered by dragging
them to the desired place.
Figure 4.2: Creating a new annotation. Annotations are created from a selection on the annotation row and can be coloured as desired.
Automated Annotation of Alignments and Groups
On loading a sequence alignment, Jalview will normally2 calculate a set of automatic annotation
rows which are shown below the alignment. For nucleotide sequence alignments, only an alignment
consensus row will be shown, but for amino acid sequences, alignment quality (based on BLOSUM
62) and physicochemical conservation will also be shown. Conservation is calculated according to
Livingstone and Barton3 . Consensus is the modal residue (or + where there is an equal top residue).
The inclusion of gaps in the consensus calculation can be toggled by right-clicking on the Consensus
label and selecting Ignore Gaps in Consensus from the pop-up context menu located with consensus
annotation row. Quality is a measure of the inverse likelihood of unfavourable mutations in the
alignment. Further details on these calculations can be found in the on-line documentation.
1 When annotating a block of positions, the text can be partly obscured by the selection highlight. Pressing the [ESC]
key clears the selection and the label is then visible.
2 Automatic annotation can be turned off in the Visual tab in the Tools ⇒ Preferences dialog box.
3 “Protein Sequence Alignments: A Strategy for the Hierarchical Analysis of Residue Conservation." Livingstone C.D.
and Barton G.J. (1993) CABIOS 9, 745-756
Exercise 14: Annotating Alignments
14.a. Load the alignment at http://www.jalview.org/tutorial/alignment.fa. Right-click on
the Conservation annotation row to bring up the context menu and select Add New
Row. A dialog box will appear asking for Annotation Name and Annotation Description. Enter “Iron binding site" and click OK. A new, empty, row appears.
14.b. Navigate to column 97. Move down and on the new annotation row called “Iron binding site, select column 97. Right click at this selection and select Label from the
context menu. Enter “Fe" in the box and click OK. Right-click on the selection again
and select Colour. Choose a colour from the colour chooser dialog and click OK. Press
[ESC] to remove the selection.
Note: depending on your Annotation sort settings, your newly created annotation row
might ’jump’ to the top or bottom of the annotation panel. Just scroll up or down to
find it again - the column you marked will still be selected.
14.c. Select columns 70-77 on the annotation row. Right-click and choose Sheet from the
context menu. You will be prompted for a label. Enter “B" and press OK. A new line
showing the sheet as an arrow appears. The colour of the label can be changed but
not the colour of the sheet arrow.
14.d. Right click on the title text of annotation row that you just created. Select Export
Annotation in context menu and, in the Export Annotation dialog box that will open,
select the Jalview format and click the [To Textbox] button.
The format for this file is given in the Jalview help. Press [F1] to open it, and find
the “Annotations File Format” entry in the “Alignment Annotations” section of the
contents pane.
14.e. Export the file to a text editor and edit the file to change the name of the annotation
row. Save the file and drag it onto the alignment view.
14.f. Add an additional helix somewhere along the row by editing the file and re-importing
Hint: Use the Export Annotation function to view what helix annotation looks like in
a Jalview annotation file.
14.g. Use the Alignment Window ⇒ File ⇒ Export Annotations... function to export all the
alignment’s annotation to a file.
14.h. Open the exported annotation in a text editor, and use the Annotation File Format
documentation to modify the style of the Conservation, Consensus and Quality annotation rows so they appear as several lines on a single line graph.
Hint: You need to change the style of annotation row in the first field of the annotation
row entry in the file, and create an annotation row grouping to overlay the three
quantitative annotation rows.
14.i. Homework for after you have completed exercise 27:
Recover or recreate the secondary structure predictions that you made from JPred.
Use the File ⇒ Export Annotation function to view the Jnet secondary structure
prediction annotation row.
Note the SEQUENCE_REF statements surrounding the row specifying the sequence
association for the annotation.
Importing Features from Databases
Jalview supports feature retrieval from public databases. It includes built in parsers for Uniprot and
ENA (or EMBL) records retrieved from the EBI. Sequences retrieved from these sources using the
sequence fetcher (see Section 1.4.5) will already possess features.
Sequence Database Reference Retrieval
Jalview maintains a list of external database references for each sequence in an alignment. These
are listed in a tooltip when the mouse is moved over the sequence ID when the View ⇒ Sequence ID
Tooltip ⇒ Show Database Refs option is enabled. Sequences retrieved using the sequence fetcher
will always have at least one database reference, but alignments imported from an alignment file
generally have no database references.
Database References and Sequence Coordinate Systems
Jalview displays features in the local sequence’s coordinate system which is given by its ‘start’ and
‘end’. Any sequence features on the sequence will be rendered relative to the sequence’s start position. If the start/end positions do not match the coordinate system from which the features were
defined, then the features will be displayed incorrectly.
Viewing and Exporting a Sequence’s Database Annotation
You can export all the database cross references and annotation terms shown in the sequence ID
tooltip for a sequence by right-clicking and selecting the [Sequence ID] ⇒ Sequence details . . . option
from the popup menu. A similar option is provided in the Selection sub-menu allowing you to obtain
annotation for the sequences currently selected.
The Sequence Details . . . option will open a window
containing the same text as would be shown in the
tooltip window, including any web links associated
with the sequence. The text is HTML, and options
on the window allow the raw code to be copied and
pasted into a web page.
Automatically Discovering a Sequence’s Database References
Jalview includes a function to automatically verify and update each sequence’s start and end numbering against any of the sequence databases that the Sequence Fetcher has access to. This function
is accessed from the Webservice ⇒ Fetch DB References sub-menu in the Alignment window. This
menu allows you to query either the set of Standard Databases, which includes EMBL, Uniprot, the
PDB, or just a specific datasource from one of the submenus. When one of the entries from this menu
is selected, Jalview will use the ID string from each sequence in the alignment or in the currently selected set to retrieve records from the external source. Any sequences that are retrieved are matched
against the local sequence, and if the local sequence is found to be a sub-sequence of the retrieved
sequence then the local sequence’s start/end numbering is updated. A new database reference mapping is created, mapping the local sequence to the external database, and the local sequence inherits
any additional annotation retrieved from the database sequence.
The database retrieval process terminates when a valid mapping is found for a sequence, or if all
database queries failed to retrieve a matching sequence. Termination is indicated by the disappearance of the moving progress indicator on the alignment window. A dialog box may be shown once
it completes which lists sequences for which records were found, but the sequence retrieved from
the database did not exactly contain the sequence given in the alignment (the “Sequence not 100%
match” dialog box).
The Fetch Uniprot IDs Dialog Box
If any sources are selected which refer to Uniprot coordinates as their reference system, then you
may be asked if you wish to retrieve Uniprot IDs for your sequence. Pressing OK instructs Jalview to
verify the sequences against Uniprot records retrieved using the sequence’s ID string. This operates
in much the same way as the Web Service ⇒ Fetch Database References function described in
Section 4.2.1. If a sequence is verified, then the start/end numbering will be adjusted to match the
Uniprot record.
Rate of Feature Retrieval
Feature retrieval can take some time if a large number of sources are selected and if the alignment
contains a large number of sequences. As features are retrieved, they are immediately added to the
current alignment view. The retrieved features are shown on the sequence and can be customised as
described previously.
4.2.2 Colouring Features by Score or Description Text
Sometimes, you may need to visualize the differences in information carried by sequence features of
the same type. This is most often the case when features of a particular type are the result of a specific type of database query or calculation. Here, they may also carry information within their textual
description, or most commonly for calculations, a score related to the property being investigated.
Jalview can shade sequence features using a graduated colourscheme in order to highlight these
variations. In order to apply a graduated scheme to a feature type, select the ‘Graduated colour’ entry in the Sequence Feature Type’s popup menu, which is opened by right-clicking the Feature Type
or Color in the Sequence Feature Settings dialog box. Two types of colouring styles are currently
supported: the default is quantitative colouring, which shades each feature based on its score, with
the highest scores receiving the ‘Max’ colour, and the lowest scoring features coloured with the ‘Min’
colour. Alternately, you can select the ‘Colour by label’ option to create feature colours according to
the description text associated with each feature. This is useful for general feature types - such as
Uniprot’s ‘DOMAIN’ feature - where the actual type of domain is given in the feature’s description.
Graduated feature colourschemes can also be used to exclude low or high-scoring features from the
alignment display. This is done by choosing your desired threshold type (either above or below),
using the drop-down menu in the dialog box. Then, adjust the slider or enter a value in the text box
to set the threshold for displaying this type of feature.
The feature settings dialog box allows you to toggle between a graduated and simple feature colourscheme
using the pop-up menu for the feature type. When a graduated scheme is applied, it will be indicated
in the colour column for that feature type - with coloured blocks or text to indicate the colouring style
and a greater than (>) or less than (<) symbol to indicate when a threshold has been defined.
Using Features to Re-order the Alignment
The presence of sequence features on certain sequences or in a particular region of an alignment can
quantitatively identify important trends in the aligned sequences. In this case, it is more useful to
re-order the alignment based on the number of features or their associated scores, rather than simply
re-colour the aligned sequences. The sequence feature settings dialog box provides two buttons: ‘Seq
sort by Density’ and ‘Seq sort by Score’, that allow you to reorder the alignment according to the
number of sequence features present on each sequence, and also according to any scores associated
with a feature. Each of these buttons uses the currently displayed features to determine the ordering,
but if you wish to re-order the alignment using a single type of feature, then you can do this from
the Feature Type’s popup menu. Simply right-click the type’s style in the Sequence Feature Settings
dialog box, and select one of the Sort by Score and Sort by Density options to re-order the alignment.
Finally, if a specific region is selected, then only features found in that region of the alignment will
be used to create the new alignment ordering.
Creating Sequence Features
Sequence features can be created simply by selecting the area in a sequence (or sequences) to form
the feature and selecting Selection ⇒ Create Sequence Feature from the right-click context menu
(Figure 4.3). A dialog box allows the user to customise the feature with respect to name, group, and
colour. The feature is then associated with the sequence. Moving the mouse over a residue associated
with a feature brings up a tool tip listing all features associated with the residue.
Figure 4.3: Creating sequence features. Features can readily be created from selections via
the context menu and are then displayed on the sequence.
Creation of features from a selection spanning multiple sequences results in the creation of one
feature per sequence. Each feature remains associated with its own sequence.
Customising Feature Display
Feature display can be toggled on or off by selecting the View ⇒ Show Sequence Features menu
option. When multiple features are present it is usually necessary to customise the display. Jalview
allows the display, colour, rendering order and transparency of features to be modified via the View
⇒ Feature Settings. . . menu option. This brings up a dialog window (Figure 4.5) which allows the
visibility of individual feature types to be selected, colours changed (by clicking on the colour of each
sequence feature type) and the rendering order modified by dragging feature types to a new position
in the list. Dragging the slider alters the transparency of the feature rendering. The Feature Settings
dialog also includes functions for more advanced feature shading schemes and buttons for sorting
the alignment according to the distribution of features. These capabilities are described further in
sections 4.2.2 and 4.2.3.
Figure 4.4: Multiple sequence features. An alignment with JPred secondary structure prediction annotation below it, and many sequence features overlaid onto the aligned
sequences. The tooltip lists the features annotating the residue below the mousepointer.
Figure 4.5: Customising sequence features. Features can be recoloured, switched on or off
and have the rendering order changed.
Sequence Feature File Formats
Jalview supports the widely used GFF tab delimited format4 and its own Jalview Features file format
for the import of sequence annotation. Features and alignment annotation are also extracted from
other formats such as Stockholm, and AMSA. URL links may also be attached to features. See the
online documentation for more details of the additional capabilities of the Jalview features file.
Exercise 15: Creating Features
15.a. Open the alignment at http://www.jalview.org/tutorial/alignment.fa. We know that
the Cysteine residues at columns 97, 102, 105 and 135 are involved in iron binding
so we will create them as features. Navigate to column 97, sequence 1. Select the
entire column by clicking in the ruler bar. Then right-click on the selection to bring
up the context menu and select Selection ⇒ Create Sequence Feature. A dialog box
will appear.
15.b. Enter a suitable Sequence Feature Name (e.g. “Iron binding site") in the appropriate
box. Click on the Feature Colour bar to change the colour if desired, add a short
description (“One of four Iron binding Cysteines") and press OK. The features will
then appear on the sequences.
15.c. Roll the mouse cursor over the new features. Note that the position given in the
tool tip is the residue number, not the column number. To demonstrate that there is
one feature per sequence, clear all selections by pressing [ESC] then insert a gap in
sequence 3 at column 95. Roll the mouse over the features and you will see that the
feature has moved with the sequence. Delete the gap you created.
15.d. Add a similar feature to column 102. When the feature dialog box appears, clicking
the Sequence Feature Name box brings up a list of previously described features.
Using the same Sequence Feature Name allows the features to be grouped.
15.e. Select View ⇒ Feature Settings. . . from the alignment window menu. The Sequence
Feature Settings window will appear. Move this so that you can see the features you
have just created. Click the check box for “Iron binding site" under Display and note
that display of this feature type is now turned off. Click it again and note that the
features are now displayed. Close the sequence feature settings box by clicking OK
or Cancel.
4 see http://www.sanger.ac.uk/resources/software/gff/spec.html
Chapter 5
Multiple Sequence Alignment
Sequences can be aligned using a range of algorithms provided by JABA web services, including
ClustalW1 , Muscle2 , MAFFT3 , ProbCons,4 T-COFFEE5 and Clustal Omega.6 Of these, T-COFFEE
is slow but accurate. ClustalW is historically the most widely used. Muscle is fast and probably best
for smaller alignments. MAFFT is probably the best for large alignments, however Clustal Omega,
released in 2011, is arguably the fastest and most accurate tool for protein multiple alignment.
Performing a multiple sequence alignment
To run an alignment web service, select the appropriate method from the Web Service ⇒ Alignment
⇒ . . . submenu (Figure 5.1). For each service you may either perform an alignment with default
settings, use one of the available presets, or customise the parameters with the ‘Edit and Run ..’
dialog box. Once the job is submitted, a progress window will appear giving information about the
job and any errors that occur. After successful completion of the job, a new alignment window is
opened with the results, in this case an alignment. By default, the new alignment will be ordered
in the same way as the input sequences. Note: many alignment programs re-order the input during
their analysis and place homologous sequences close together, the MSA algorithm ordering can be
recovered using the ‘Algorithm ordering’ entry within the Calculate ⇒ Sort sub menu.
1 “CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting,
position specific gap penalties and weight matrix choice." Thompson JD, Higgins DG, Gibson TJ (1994) Nucleic Acids
Research 22, 4673-80
2 “MUSCLE: a multiple sequence alignment method with reduced time and space complexity" Edgar, R.C. (2004) BMC
Bioinformatics 5, 113
3 “MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform" Katoh, K., Misawa,
K., Kuma, K. and Miyata, T. (2002) Nucleic Acids Research 30, 3059-3066. and “MAFFT version 5: improvement in
accuracy of multiple sequence alignment" Katoh, K., Kuma, K., Toh, H. and Miyata, T. (2005) Nucleic Acids Research 33,
4 PROBCONS: Probabilistic Consistency-based Multiple Sequence Alignment. Do, C.B., Mahabhashyam, M.S.P.,
Brudno, M., and Batzoglou, S. (2005) Genome Research 15 330-340.
5 T-Coffee: A novel method for multiple sequence alignments. (2000) Notredame, Higgins and Heringa JMB 302 205-217
6 Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Sievers F, Wilm
A, Dineen DG, Gibson TJ, Karplus K, Li W, Lopez R, McWilliam H, Remmert M, Soding J, Thompson JD, Higgins DG
(2011) Molecular Systems Biology 7 539 doi:10.1038/msb.2011.75
Realignment to add sequences to an existing alignment
The re-alignment option is currently only supported by Clustal Omega and ClustalW. When performing a re-alignment, Jalview submits the current selection to the alignment service complete with any
existing gaps. Realignment with ClustalW is useful when one wishes to align additional sequences
to an existing alignment without any further optimisation to the existing alignment. ClustalO’s realignment works by generating a probabilistic model (a.k.a HMM) from the original alignment, and
then realigns all sequences to this profile. For a well aligned MSA, this process will simply reconstruct the original alignment (with additonal sequences), but in the case of low quality MSAs, some
differences may be introduced.
Figure 5.1: Multiple alignment via web services The appropriate method is selected from the
menu (left), a status box appears (centre), and the results appear in a new window
Alignments of Sequences that include Hidden Regions
If the view or selected region submitted for alignment contains hidden regions, then only the visible
sequences will be submitted to the service. Furthermore, each contiguous segment of sequences
will be aligned independently (resulting in a number of alignment ‘subjobs’ appearing in the status
window). Finally, the results of each subjob will be concatenated with the hidden regions in the input
data prior to their display in a new window. This approach ensures that 1) hidden column boundaries
in the input data are preserved in the resulting alignment - in a similar fashion to the constraint that
hidden columns place on alignment editing (see Section 2.6.2 and 2) hidden columns can be used to
preserve existing parts of an alignment whilst the visible parts are locally refined.
Alignment Service Limits
Multiple alignment is a computationally intensive calculation. Some JABA server services and service presets only allow a certain number of sequences to be aligned. The precise number will depend
on the server that you are using to perform the alignment. Should you try to submit more sequences
than a service can handle, then an error message will be shown informing you of the maximum
number allowed by the server.
Exercise 16: Multiple Sequence Alignment
16.a. Close all windows and open the alignment at http://www.jalview.org/tutorial/unaligned.fa.
Select Web Service ⇒ Alignment ⇒ Muscle with Defaults. A window will open giving
the job status. After a short time, a second window will open with the results of the
16.b. Return to the first sequence alignment window by clicking on the window, and repeat
using ClustalO (Omega) and MAFFT, from the Web Service ⇒ Alignment menu, using the same initial alignment. Compare them and you should notice small differences.
16.c. Select the last three sequences in the MAFFT alignment, and de-align them with
Edit ⇒ Remove All Gaps. Press [ESC] to deselect these sequences. Then submit this
view for re-alignment with ClustalO.
16.d. Return to the alignment window in section (c), use [CTRL]-Z (undo) to recover the
alignment of the last three sequences in this MAFFT alignment. Once the ClustalO
re-alignment has completed, compare the results of re-alignment of the three sequences with their alignment in the original MAFFT result.
16.e. Select columns 60 to 125 in the original MAFFT alignment and hide them, by right
clicking the mouse to bring up context menu. Select Web Service ⇒ Alignment ⇒
Mafft with Defaults to submit the visible portion of the alignment to MAFFT. When
the web service job pane appears, note that there are now two alignment job status
panes shown in the window.
16.f. When the MAFFT job has finished, compare the alignment of the N-terminal visible
region in the result with the corresponding region of the original alignment.
16.g. If you wish, select and hide a few more columns in the N-terminal region, and submit
the alignment to the service again and explore the effect of local alignment on the
non-homologous parts of the N-terminal region.
See the video at: http://www.jalview.org/training/Training-Videos.
Customising the Parameters used for Alignment
JABA web services allow you to vary the parameters used when performing a bioinformatics analysis. For JABA alignment services, this means you are usually able to modify the following types of
• Amino acid or nucleotide substitution score matrix
• Gap opening and widening penalties
• Types of distance metric used to construct guide trees
• Number of rounds of re-alignment or alignment optimisation
Figure 5.2: Jalview’s JABA alignment service parameter editing dialog box.
Getting Help on the Parameters for a Service
Each parameter available for a method usually has a short description, which Jalview will display
as a tooltip, or as a text pane that can be opened under the parameter’s controls. In the parameter
shown in Figure 5.3, the description was opened by selecting the button on the left hand side. Online
help for the service can also be accessed, by right clicking the button and selecting a URL from the
pop-up menu that will open.
Figure 5.3: ClustalW parameter slider detail. From the ClustalW Clustal ⇒ Edit settings
and run ... dialog box.
Alignment Presets
The different multiple alignment algorithms available from JABA vary greatly in the number of
adjustable parameters, and it is often difficult to identify what are the best values for the sequences
that you are trying to align. For these reasons, each JABA service may provide one or more presets
– which are pre-defined sets of parameters suited for particular types of alignment problem. For
instance, the Muscle service provides the following presets:
• Large alignments (balanced)
• Protein alignments (fastest speed)
• Nucleotide alignments (fastest speed)
The presets are displayed in the JABA web services submenu, and can also be accessed from the
parameter editing dialog box, which is opened by selecting the ‘Edit settings and run ...’ option from
the web services menu. If you have used a preset, then it will be mentioned at the beginning of the
job status file shown in the web service job progress window.
5.2.3 User Defined Presets
Jalview allows you to create your own presets for a particular service. To do this, select the ‘Edit
settings and run ...’ option for your service, which will open a parameter editing dialog box like the
one shown in Figure 5.2.
The top row of this dialog allows you to browse the existing presets, and when editing a parameter
set, allows you to change its nickname. As you adjust settings, buttons will appear at the top of the
parameters dialog that allow you to Revert or Update the currently selected user preset with your
changes, Delete the current preset, or Create a new preset, if none exists with the given name. In
addition to the parameter set name, you can also provide a short description for the parameter set,
which will be shown in the tooltip for the parameter set’s entry in the web services menu.
Saving Parameter Sets
When creating a custom parameter set, you will be asked for a file name to save it. The location of the
file is recorded in the Jalview user preferences in the same way as a custom alignment colourscheme,
so when Jalview is launched again, it will show your custom preset amongst the options available
for running the JABA service.
Protein Alignment Conservation Analysis
The Web Service ⇒ Conservation menu controls the computation of up to 17 different amino acid conservation measures for the current alignment view. The JABAWS AACon Alignment Conservation
Calculation Service, which is used to calculate these scores, provides a variety of standard measures
described by Valdar in 20027 as well as an efficient implementation of the SMERFs score developed
by Manning et al. in 2008.8
5.3.1 Enabling and Disabling AACon Calculations
When the AACon Calculation entry in the Web Service ⇒ Conservation menu is ticked, AACon
calculations will be performed every time the alignment is modified. Selecting the menu item will
enable or disable automatic recalculation.
7 Scoring residue conservation. Valdar (2002) Proteins: Structure, Function, and Genetics 43 227-241.
8 SMERFS Score Manning et al. BMC Bioinformatics 2008, 9 51 doi:10.1186/1471-2105-9-51
5.3.2 Configuring which AACon Calculations are Performed
The Web Service ⇒ Conservation ⇒ Change AACon Settings ... menu entry will open a web services
parameter dialog for the currently configured AACon server. Standard presets are provided for quick
and more expensive conservation calculations, and parameters are also provided to change the way
that SMERFS calculations are performed. AACon settings for an alignment are saved in Jalview
projects along with the latest calculation results.
5.3.3 Changing the Server used for AACon Calculations
If you are working with alignments too large to analyse with the public JABAWS server, then you
will most likely have already configured additional JABAWS servers. By default, Jalview will chose
the first AACon service available from the list of JABAWS servers available. If available, you can
switch to use another AACon service by selecting it from the Web Service ⇒ Conservation ⇒ Switch
Server submenu.
Chapter 6
Analysis of Alignments
Jalview provides support for sequence analysis in two ways. A number of analytical methods are
‘built-in’, these are accessed from the Calculate alignment window menu. Computationally intensive
analyses are run outside Jalview via web services - and found under the Web Service menu. In this
section, we describe the built-in analysis capabilities common to both the Jalview Desktop and the
JalviewLite applet.
Principal components analysis calculations create a spatial representation of the similarities within
the current selection or the whole alignment if no selection has been made. After the calculation
finishes, a 3D viewer displays each sequence as a point in 3D ‘similarity space’. Sets of similar
sequences tend to lie near each other in this space. Note: The calculation is computationally expensive, and may fail for very large sets of sequences - because the JVM has run out of memory. Memory
issues, and how to overcome them, were discussed in Section 1.4.6.
What is PCA?
Principal components analysis is a technique for examining the structure of complex data sets. The
components are a set of dimensions formed from the measured values in the data set, and the principal component is the one with the greatest magnitude, or length. The sets of measurements that
differ the most should lie at either end of this principal axis, and the other axes correspond to less
extreme patterns of variation in the data set. In this case, the components are generated by an
eigenvector decomposition of the matrix formed from the sum of pairwise substitution scores at each
aligned position between each pair of sequences. The basic method is described in the 1995 paper by
G. Casari, C. Sander and A. Valencia 1 and implemented at the SeqSpace server at the EBI.
Jalview provides two different options for the PCA calculation: SeqSpace and Jalview mode. In
SeqSpace mode, PCAs are computed using the identity matrix, and gaps are treated as ’the unknown
1 Nature Structural Biology (1995) 2, 171-8. PMID: 7749921
residue’ (this actually differs from the original SeqSpace paper, and will be adjusted in a future
version of Jalview). In Jalview mode, PCAs are computed using the chosen score matrix - which for
protein sequences, defaults to BLOSUM 62, and for nucleotides, is the DNA identity matrix that also
treats Us and Ts as identical, to support analysis of both RNA and DNA alignments. The Change
Parameters allows the calculation method and score models to be changed.2
The PCA Viewer
PCA analysis can be launched from the Calculate ⇒ Principal Component Analysis menu option.
PCA requires a selection containing at least 4 sequences. A window opens containing the
PCA tool (Figure 6.1). Each sequence is represented by a small square, coloured by the background
colour of the sequence ID label. The axes can be rotated by clicking and dragging the left mouse
button and zoomed using the ↑ and ↓ keys or the scroll wheel of the mouse (if available). A tool
tip appears if the cursor is placed over a sequence. Sequences can be selected by clicking on them.
[CTRL]-Click can be used to select multiple sequences.
Labels will be shown for each sequence by toggling the View ⇒ Show Labels menu option, and
the plot background colour changed via the View ⇒ Background Colour.. dialog box. A graphical
representation of the PCA plot can be exported as an EPS or PNG image via the File ⇒ Save As ⇒
. . . submenu.
Exercise 17: Principal Component Analysis
17.a. Load the alignment at http://www.jalview.org/tutorial/alignment.fa.
17.b. Select the menu option Calculate ⇒ Principal Component Analysis. A new window
will open. Move this window within the desktop so that the tree, alignment and PCA
viewer windows are all visible. Try rotating the plot by clicking and dragging the
mouse on the plot in the PCA window. Note that clicking on points in the plot will
highlight the sequences on the alignment.
17.c. Select Calculate ⇒ Calculate Tree ⇒ Neighbour Joining Using BLOSUM62. A new
tree window will appear. Place the mouse cursor on the tree window so that the
tree partition location will divide the alignment into a number of groups, each of a
different colour. Note how the colour of the sequence ID label matches both the colour
of the partitioned tree and the points in the PCA plot.
See the video at: http://www.jalview.org/training/Training-Videos.
PCA Data Export
Although the PCA viewer supports export of the current view, the plots produced are rarely suitable for direct publication. The PCA viewer’s File menu includes a number of options for exporting
the PCA matrix and transformed points as comma separated value (CSV) files. These files can be
imported by tools such as R or gnuplot in order to graph the data.
2 See
6.2. TREES
Figure 6.1: PCA Analysis.
Figure 6.2: Calculating Trees Jalview provides a range of options for calculating trees. Jalview
can also load precalculated trees in Newick format (right).
Jalview can calculate and display trees, providing interactive tree-based grouping of sequences
though a tree viewer. All trees are calculated via the Calculate ⇒ Calculate Tree ⇒ . . . submenu.
Trees can be calculated from distance matrices determined from % identity or aggregate BLOSUM
62 score using either Average Distance (UPGMA) or Neighbour Joining algorithms. The input data
for a tree is either the selected region or the whole alignment, excluding any hidden regions.
On calculating a tree, a new window opens (Figure 6.2) which contains the tree. Various display
settings can be found in the tree window View menu, including font, scaling and label display options.
The File ⇒ Save As submenu contains options for image and Newick file export. Newick format is
a standard file format for trees which allows them to be exported to other programs. Jalview can
also read in external trees in Newick format via the File ⇒ Load Associated Tree menu option. Leaf
names on imported trees will be matched to the associated alignment - unmatched leaves will still
be displayed, and can be highlighted using the View ⇒ Mark Unlinked Leaves menu option.
Figure 6.3: Interactive Trees The tree level cutoff can be used to designate groups in Jalview.
Clicking on the tree brings up a cursor across the height of the tree. The sequences are automatically
partitioned and coloured (Figure 6.3). To group them together, select the Calculate ⇒ Sort ⇒ By Tree
Order ⇒ . . . alignment window menu option and choose the correct tree. The sequences will then
be sorted according to the leaf order currently shown in the tree view. The coloured background to
the sequence IDs can be removed with Select ⇒ Undefine Groups from the alignment window menu.
Note that tree partitioning will also remove any groups and colourschemes on a view, so create a new
view ([CTRL-T]) if you wish to preserve these.
Recovering input Data for a Tree or PCA Plot Calculation
The File ⇒ Input Data option will open a new alignment window containing
the original data used to calculate the tree or PCA plot (if available). This
function is useful when a tree has been created and then the alignment subsequently changed.
Changing the associated View for a Tree or PCA Viewer
The View ⇒ Associated Nodes With ⇒ .. submenu is shown
when the viewer is associated with an alignment that is involved in multiple views. Selecting a different view does not
affect the tree or PCA data, but will change the colouring and
display of selected sequences in the display according to the
colouring and selection state of the newly associated view.
Tree Based Conservation Analysis
Trees reflect the pattern of global sequence similarity exhibited by the alignment, or region within
the alignment, that was used for their calculation. The Jalview tree viewer enables sequences to
6.2. TREES
be partitioned into groups based on the tree. This is done by clicking within the tree viewer window. Once subdivided, the conservation between and within groups can be visually compared in
order to better understand the pattern of similarity revealed by the tree and the variation within the
clades partitioned by the grouping. The conservation based colourschemes and the group associated
conservation and consensus annotation (enabled using the alignment window’s View ⇒ Autocalculated Annotation ⇒ Group Conservation and Group Consensus options) can help when working with
larger alignments.
Exercise 18: Trees
Ensure that you have at least 1G memory available in Jalview.
(Start with link: http://www.jalview.org/services/launchApp?jvm-max-heap=1G, or in the Development section of the Jalview web site (http://www.jalview.org/development/developmentbuilds) in the table, go to “latest official build” row and “Webstart” column, click on “2G”.)
18.a. Open the alignment at http://www.jalview.org/tutorial/alignment.fa. Select Calculate
⇒ Calculate Tree ⇒ Neighbour Joining Using BLOSUM62. A tree window opens.
18.b. Click on the tree window, a cursor will appear. Note that placing this cursor divides
the tree into a number of groups by colour. Place the cursor to give about 4 groups.
18.c. In the alignment window, select Calculate ⇒ Sort ⇒ By Tree Order ⇒ Neighbour
Joining Tree using BLOSUM62 from... . The sequences are reordered to match the
order in the tree and groups are formed implicitly. Alternatively in the tree window,
select View ⇒ Sort Alignment by Tree.
18.d. Select Calculate ⇒ Calculate Tree ⇒ Neighbour Joining Using % Identity. A new
tree window will appear. The group colouring makes it easy to see the differences
between the two trees calculated by the different methods.
18.e. Select from sequence 2 column 60 to sequence 12 column 123. Select Calculate ⇒
Calculate Tree ⇒ Neighbour Joining Using BLOSUM62. A new tree window will
appear. The tree contains 11 sequences. It has been coloured according to the already
selected groups from the first tree and is calculated purely from the residues in the
Comparing the location of individual sequences between the three trees illustrates the importance of selecting appropriate regions of the alignment for the calculation of trees.
See the video at: http://www.jalview.org/training/Training-Videos.
Exercise 19: Pad Gaps in an Alignment
19.a. Open the alignment at http://www.jalview.org/tutorial/alignment.fa. In alignment
window, ensure that the Edit ⇒ Pad Gaps option is not ticked, and insert one gap
anywhere in the alignment.
19.b. Select Calculate ⇒ Calculate Tree ⇒ Neighbour Joining Using BLOSUM62.
A warning dialog box “Sequences not aligned” appears because the sequences
input to the tree calculation are of different lengths.
19.c. Select Edit ⇒ tick Pad Gaps and perform the tree calculation again. This time a new
tree should appear - because padding gaps ensures all the sequences are the same
length after editing.
Pad Gaps option can be set in Preferences using Tool ⇒ Preference ⇒ Editing.
See the video at: http://www.jalview.org/training/Training-Videos.
Figure 6.4: The Redundancy Removal dialog box opened from the edit menu. Sequences that exceed the current percentage identity threshold and are to be removed are highlighted
in black.
Exercise 20: Tree Based Conservation Analysis
20.a. Load the PF03460 PFAM seed alignment using the sequence fetcher. Select Colour
⇒ Taylor ⇒ By Conservation, set Conservation shading threshold at around 20.
20.b. Build a Neighbour joining tree using Select Calculate ⇒ Calculate Tree ⇒ Neighbour
Joining Using BLOSUM62.
20.c. Use the mouse cursor to select a point on the tree to partition the alignment into
several sections.
20.d. Select View ⇒ Sort Alignment By Tree option in the tree window to re-order the sequences in the alignment using the calculated tree. Examine the variation in colouring between different groups of sequences in the alignment window.
20.e. You may find it easier to browse the alignment if you first uncheck the Annotations
⇒ Show Annotations option. Open the Overview Window within the View menu to
aid navigation.
20.f. Try changing the colourscheme of the residues in the alignment to BLOSUM62
(whilst ensuring that Apply Colour to All Groups is selected).
Note: You may want to save the alignment and tree as a project file, since it is used in the
next set of exercises.
See the video at: http://www.jalview.org/training/Training-Videos.
6.2.2 Redundancy Removal
The redundancy removal dialog box is opened using the Edit ⇒ Remove Redundancy. . . option in
the alignment menu. As its menu option placement suggests, this is actually an alignment editing
function, but it is convenient to describe it here. The redundancy removal dialog box presents a
percentage identity slider which sets the redundancy threshold. Aligned sequences which exhibit
a percentage identity greater than the current threshold are highlighted in black. The [Remove]
button can then be used to delete these sequences from the alignment as an edit operation3 .
3 Which can usually be undone. A future version of Jalview may allow redundant sequences to be hidden, or represented
by a chosen sequence, rather than deleted.
Subdividing the Alignment According to Specific Mutations
It is often necessary to explore variations in an alignment that may correlate with mutations observed in a particular region; for example, sites exhibiting single nucleotide polymorphism, or residues
involved in substrate recognition in an enzyme. One way to do this would be to calculate a tree using
the specific region, and subdivide it in order to partition the alignment. However, calculating a tree
can be slow for large alignments, and the tree may be difficult to partition when complex mutation
patterns are being analysed. The Select ⇒ Make groups for selection function was introduced to
make this kind of analysis easier. When selected, it will use the characters in the currently selected
region to subdivide the alignment. For example, if a single column is selected, then the alignment (or
each group defined on the alignment) will be divided into groups based on the residue or nucleotide
found at that position. These new groups are annotated with the characters in the selected region,
and Jalview’s group based conservation analysis annotation and colourschemes can then be used to
reveal any associated pattern of sequence variation across the whole alignment.
Pairwise Alignments
Jalview can calculate optimal pairwise alignments between arbitrary sequences via the Calculate ⇒
Pairwise Alignments. . . menu option. Global alignments of all pairwise combinations of the selected
sequences are performed and the results returned in a text box.
Exercise 21: Remove Redundant Sequences
21.a. Using the alignment generated in the previous exercise (exercise 20). In the alignment window, you may need to deselect groups using Esc key.
21.b. In the Edit menu select Remove Redundancy to open the Redundancy threshold selection dialog. Adjust the redundancy threshold value, start at 50 and increase the
value to 65. Sequences selected will change colour in the Sequence ID panel. Select “Remove” to remove the sequences that are more than 65% similar under this
21.c. From the tree window, select View ⇒ Mark Unlinked Leaves option, and note that
the removed sequences are now prefixed with a * in the tree view.
21.d. Use the [Undo] button in the Redundancy threshold selection dialog box to recover
the sequences. Note that the * symbols disappear from the tree display.
21.e. Experiment with the redundancy removal and observe the relationship between the
percentage identity threshold and the pattern of unlinked nodes in the tree display.
Figure 6.5: Pairwise alignment of sequences. Pairwise alignments of three selected sequences are shown in a textbox.
Exercise 22: Group Conservation Analysis
22.a. Re-use or recreate the alignment and tree which you worked with in the tree based
conservation analysis exercise (exercise 20).
22.b. In the View menu in the alignment window, select New View to create a new view.
Ensure the annotation panel is displayed (Show annotation in Annotations menu).
Enable the display of Group Consensus option by checking Group Consensus in the
Annotation ⇒ Autocalculated Annotation submenu in the alignment window.
22.c. Displaying the sequence logos will make it easier to see the different residue populations within each group. Activate the logo by right clicking on the Consensus
annotation row to open the context menu and select the Show Logo option.
22.d. In the column alignment ruler, select a column exhibiting about 50% conservation
that lies within the central conserved region of the alignment. (Column 74 is used in
the Tree video).
22.e. Subdivide the alignment according to this selection using Select ⇒ Make groups for
22.f. Re-order the alignment according to the new groups that have been defined by selecting Calculate ⇒ Sort ⇒ By Group.
Click on the group annotation row IDs to select groups exhibiting a specific mutation.
22.g. Select another column exhibiting about 50% conservation overall, and subdivide the
alignment further. Note that the new groups inherit the names of the original groups,
allowing you to identify the combination of mutations that resulted in the subdivision.
22.h. Clear the groups, and try to subdivide the alignment using two non-adjacent columns.
Hint: You may need to hide the intervening columns before you can select both of the
columns that you wish to use to subdivide the alignment.
22.i. Switch back to the original view, and experiment with subdividing the tree groups
made in the previous exercise.
See the video at: http://www.jalview.org/training/Training-Videos.
Chapter 7
Working with 3D structures
Jalview facilitates the use of 3D structure data for the analysis of alignments by providing a linked
view of structures associated with the aligned sequences. It also allows sequence, secondary structure and B-factor data to be imported from structure files, and supports the use of the EMBL-EBI’s
SIFTS database to construct accurate mappings between UniProt protein sequences and structures
retrieved from the PDB.
Molecular graphics systems supported by Jalview
Jalview can interactively view 3D structure using Jmol, a Java based molecular viewing program1
integrated with Jalview.2 It also supports the use of UCSF Chimera, a powerful molecular graphics
system that needs separate installation. Jalview can also read PDB and mmCIF format files directly
to extract sequences and secondary structure information, and retrieve records from the European
Protein Databank (PDBe) using the Sequence Fetcher (see 1.4.5).
7.1.1 Configuring the default structure viewer
To configure which viewer is used when creating a new structure view, open the Structures preferences window via Tools ⇒ Preferences. . . and select either JMOL or CHIMERA as the default viewer.
If you select Chimera, Jalview will search for the installed program, and if it cannot be found, you
will be prompted to locate the Chimera binary, or alternately, open the UCSF Chimera download
page to obtain the software.
1 See the Jmol homepage
http://www.jmol.org for more information.
2 Earlier versions of Jalview included MCView - a simple main chain structure viewer. Structures are visualized as an
alpha carbon trace and can be viewed, rotated and coloured using the sequence alignment.
Automatic Association of PDB Structures with Sequences
Jalview will attempt to automatically determine which structures are associated with a sequence via
its ID, and any associated database references. To do this for a particular sequence or the current
selection, open the Sequence ID popup menu and select View 3D Structure, to open the 3D Structure
When the structure chooser is first opened, if no database identifiers are available, Jalview will automatically perform a database reference retrieval (See 4.2.1) to discover identifiers for the sequences
to use to search the PDB. This can take a few seconds for each sequence and will be performed for all
selected sequences.3
Once the retrieval has finished, the structure chooser dialog will show any available PDB entries for
the selected sequences.
7.2.1 Drag-and-Drop Association of PDB Files with Sequences by Filename Match
If you have PDB files stored on your computer named the same way as the sequences in the alignment, then you can drag them from their location on the file browser onto an alignment window.
Jalview will search the alignment for sequences with IDs that match any of the files, and offer a
dialog like the one in Figure 7.1.
If no associations are made, then sequences extracted from the structure will be simply added to the
alignment. However, if only some of the PDB files are associated, Jalview will raise another dialog
box giving you the option to add any remaining sequences from the PDB structure files not present
in the alignment. This allows you to easily decorate sequences in a newly imported alignment with
any corresponding structures you’ve already collected in a directory accessible from your computer.4
After associating sequencesÂăwith PDB files, you can view the PDB structures by opening the Sequence ID popup menu and selecting View 3D Structure. The PDB files you loaded will be shown in
the Cached Structures view, after selecting it from the drop down menu in the dialog box.
Viewing Structures
The structure viewer is launched via the Sequence ID context menu. To view structures associated
with a sequence or a selected set of sequences in the alignment, simply right click the mouse to open
the context menu, and select 3D Structure data . . . to open the Structure Chooser dialog box.
If any of the currently selected sequences have structures in the PDB, they will appear in the
Structure Chooser dialog box. The structures can be ranked by different parameters, but are by
3 After this is done, you can can see the added database references in a tool tip by mousing over the sequence ID. You
can use the View ⇒ Sequence ID Tooltip ⇒ Show Db References submenu option to enable or disable these data in the
4 We plan to extend this facility in future so Jalview will automatically search for PDB files matching your sequence
within a local directory. Check out Jalview issue 801
Figure 7.1: Associating PDB files with sequences by drag-and-drop. Dragging PDB files
onto an alignment of sequences with names matching the dragged files names (A),
results in a dialog box (B) that gives the option to associate each file with any sequences with matching IDs.
default ordered according to their PDB quality score.
To view one or more structures, simply click View to open a structure viewer containing the structures selected in the dialog. If several structures were picked, these will be shown superposed according to the alignment. You may find Jalview has already picked the best structure - using one of
the criteria shown in the dropdown menu (e.g. ’Best Quality’, which is picked by default). However,
you are free to select your own.
The structure(s) to be displayed will be downloaded or loaded from the local file system, and shown
as a ribbon diagram coloured according to the associated sequence in the current alignment view
(Figure 7.2 (right)). The structure can be rotated by clicking and dragging in the structure window.
The structure can be zoomed using the mouse scroll wheel or by [SHIFT]-dragging the structure.
Moving the mouse cursor over a sequence to which the structure is linked in the alignment view
highlights the respective residue’s sidechain atoms. The sidechain highlight may be obscured by
other parts of the molecule. Similarly, moving the cursor over the structure shows a tooltip and
highlights the corresponding residue in the alignment. Clicking the alpha carbon or phosphorous
backbone atom will toggle the highlight and residue label on and off. Often, the position highlighted
in the sequence may not be in the visible portion of the current alignment view and the sliders will
scroll automatically to show the position. If the alignment window’s View ⇒ Automatic Scrolling
option is not selected, however, then the automatic adjustment will be disabled for the current view.
Customising Structure Display
Structure display can be modified using the Colour and View menus in the structure viewer. The
background colour can be modified by selecting the Colours ⇒ Background Colour. . . option.
By default, the structure will be coloured in the same way as the associated sequence(s) in the align-
Figure 7.2: Structure visualization Structure viewers are launched from the 3D Structure
chooser dialog (left). Jalview shows the displayed structures coloured according the
alignment view (right).
ment view from which it was launched. The structure can be coloured independently of the sequence
by selecting an appropriate colour scheme from the Colours menu. It can be coloured according to
the alignment using the Colours ⇒ By Sequence option. The image in the structure viewer can be
saved as an EPS or PNG with the File ⇒ Save As ⇒ . . . submenu, which also allows the raw data to
be saved as PDB format. The mapping between the structure and the sequence (how well and which
parts of the structure relate to the sequence) can be viewed with the File ⇒ View Mapping menu
Using the Jmol Visualization Interface
Jmol has a comprehensive set of selection and visualization functions that are accessed from the
Jmol popup menu (by right-clicking in the Jmol window or by clicking the Jmol logo). Molecule
colour and rendering style can be manipulated, and distance measurements and molecular surfaces
can be added to the view. It also has its own “Rasmol5 -like” scripting language, which is described
elsewhere6 . Jalview utilises the scripting language to interact with Jmol and to store the state of
a Jmol visualization within Jalview archives, in addition to the PDB data file originally loaded or
retrieved by Jalview. To access the Jmol scripting environment directly, use the Jmol ⇒ Console
menu option.
If you would prefer to use Jmol to manage structure colours, then select the Colours ⇒ Colour with
Jmol option. This will disable any automatic application of colour schemes when new structure data
is added, or when associated alignment views are modified.
5 See
Jmol Scripting reference: http://www.stolaf.edu/academics/chemapps/jmol/docs/
6 Jmol Wiki:
Exercise 23: Viewing Structures with the integrated Jmol Viewer
23.a. Load the alignment at http://www.jalview.org/examples/exampleFile.jar.
23.b. Right-click on the sequence ID label of FER1_SPIOL to open the ID popup menu and
select 3D Structure. After a short pause, a Structure Chooser dialog will open for the
sequence, listing available structure data from the PDB. Select 1A70 from the list
and click View.
The Structure Chooser dialog presents available PDB structures by querying the
EMBL-EBI’s PDBe web API. Extra information can be including in this window by
checking boxes in the columns of the “Customise Displayed Options” tab.
23.c. By default the Jmol structure viewer opens in the Jalview desktop. Rotate the
molecule by clicking and dragging in the structure viewing box. Zoom with the mouse
scroll wheel.
23.d. Roll the mouse cursor along the FER1_SPIOL sequence in the alignment. Note that
if a residue in the sequence maps to one in the structure, a label will appear next to
that residue in the structure viewer.
23.e. Move the mouse over the structure. In the Jmol viewer, placing the mouse over a
part of the structure will bring up a tool tip indicating the name and number of
that residue. In the alignment window, the corresponding residue in the sequence is
highlighted in black.
23.f. Clicking the alpha carbon toggles the highlight and residue label on and off. Try this
by clicking on a set of three or four adjacent residues so that the labels are persistent,
then finding where they are in the sequence.
23.g. In the structure viewer menu, select Colours ⇒ Background Colour. . . and choose a
suitable colour. Press OK to apply this.
23.h. Select File ⇒ Save As ⇒ PNG and save the image. On your computer, view this with
a suitable program.
23.i. Select File ⇒ View Mapping from the structure viewer menu. A new window opens
showing the residue by residue alignment between the sequence and the structure.
23.j. Select File ⇒ Save ⇒ PDB file and choose a new filename to save the PDB file. Once
the file is saved, open the location in your file browser (or explorer window) and drag
the PDB file that you just saved on to the Jalview desktop (or load it from the Jalview
Desktop ⇒ Input Alignment ⇒ From File menu). Verify that you can open and view
the associated structure from the sequence ID context menu’s 3D Structure submenu
in the new alignment window.
23.k. In the Jmol window, right click on the structure window and explore the menu options. Try to change the style of molecular display - for example by using the Jmol ⇒
Select (n) ⇒ All command (where n is the number of residues selected), and then the
Jmol ⇒ Style ⇒ Scheme ⇒ Ball and Stick command.
23.l. In the alignment window, use the File ⇒ Save As.. function to save the alignment
as a Jalview Project. Now close the alignment and the structure view, and load the
project file you just saved. Verify that the Jmol display is as it was when you just
saved the file.
See the video at: http://www.jalview.org/training/Training-Videos.
Exercise 24: Setting Chimera as the default 3D Structure Viewer
Jalview supports molecular structure visualization using both Jmol and Chimera 3D viewers. Jmol is the default viewer, however Chimera can be set up as the default choice from
24.a. First, Chimera must be downloaded and installed on the computer. Chimera program
is available on the UCSF web site https://www.cgl.ucsf.edu/chimera/download.html.
24.b. In the desktop menu, select Tool ⇒ Preferences. In the “Structure” tab set Default
structure viewer as Chimera; then click OK.
24.c. Close the Jalview program, from the Desktop menu select Jalview ⇒ Quit Jalview.
Then reopen Jalview, Chimera should open as the default viewer.
Note: The Jmol structure viewer sits within the Jalview desktop. However the Chimera
structure viewer sits outside the Jalview desktop and a Chimera view window sits inside the
Jalview desktop.
See the video at: http://www.jalview.org/training/Training-Videos.
7.3.2 Superimposing Structures
Many comparative biomolecular analysis investigations aim to determine if the biochemical properties of a given molecule are significantly different to its homologues. When structure data is
available, comparing the shapes of molecules by superimposing them enables substructure that may
impart different behaviour to be quickly identified. The identification of optimal 3D superposition
involves aligning 3D data rather than sequence symbols, but the result can still be represented as a
sequence alignment, where columns indicate positions in each molecule that should be superposed
to recreate the optimal 3D alignment.
Jalview can employ Jmol’s 3D fitting routines7 to recreate 3D structure superpositions based on the
correspondences defined by one or more sequence alignments involving structures shown in the Jmol
display. Superposition based on the currently displayed alignment view happens automatically if a
structure is added to an existing Jmol display using the 3D Structure option in the Sequence ID
popup menu to open the Structure Chooser dialog box. Select the structures required and select
View. A new Jmol view opens containing superposed structures if the current selection contains two
or more sequences with associated structures.
Obtaining the RMSD for a Superposition
The RMSD (Root Mean Square Deviation) is a measure of how similar the structures are when they
are superimposed. Figure 7.3 shows a superposition created during the course of Exercise 25. The
parts of each molecule used to construct the superposition are rendered using the cartoon style, with
other parts of the molecule drawn in wireframe. The Jmol console, which has been opened after the
superposition was performed, shows the RMSD report for the superposition. Full information about
the superposition is also reported on the Jalview console.8 This output also includes the precise atom
pairs used to superpose structures.
7 See http://chemapps.stolaf.edu/jmol/docs/?ver=12.2#compare for more information.
8 The Jalview Java Console is opened from Tools ⇒ Java Console option in the Desktop’s menu bar
Choosing which part of the Alignment is used for Structural Superposition
Jalview uses the visible part of each alignment view to define which parts of each molecule are to
be superimposed. Hiding a column in a view used for superposition will remove that correspondence
from the set, and will exclude it from the superposition and RMSD calculation. This allows the
selection of specific parts of the alignment to be used for superposition. Only columns that define a
complete set of correspondences for all structures will be used for structural superposition, and as a
consequence, the RMSD values generated for each pair of structures superimposed can be directly
In order to recompute a superposition after changing a view or editing the alignment, select the
Jmol ⇒ Align Structures menu option. The Jmol ⇒ Superpose with .. submenu allows you to choose
which of the associated alignments and views are to be used to create the set of correspondences.
This menu is useful when composing complex superpositions involving multi-domain and multichain complexes, when correspondences may be defined by more than one alignment.
Note that these menu options appear when you have two or more structures in one Jmol viewer.
Figure 7.3: Superposition of two ferredoxin structures. The alignment on the left was
used by Jalview to superpose structures associated with the FER1_SPIOL and
FER1_MAIZE sequences in the alignment. Parts of each structure used for superposition are rendered as a cartoon, the remainder rendered in wireframe. The RMSD
between corresponding positions in the structures before and after the superposition
is shown in the Jmol console.
Colouring Structure Data Associated with Multiple Alignments and Views
Normally, the original view from which a particular structure view was opened will be the one used
to colour structure data. If alignments involving sequences associated with structure data shown in
a Jmol have multiple views, Jalview gives you full control over which alignment, or alignment view,
is used to colour the structure display. Sequence-structure colouring associations are changed via the
View ⇒ Colour by .. menu, which lists all views associated with data shown in the embedded Jmol
view. A tick is shown beside views currently used as colouring source, and moving the mouse over
each view will bring it to the front of the alignment display, allowing you to browse available colour
sources prior to selecting one. If the Select many views option is selected, then multiple views can be
selected as sources for colouring the structure data. Invert selection and Select all views options are
also provided to quickly change between multi-view selections.
Note that the Select many views option is useful if you have different views that colour different
areas or domains of the alignment. This option is further explored in Section 25.
Exercise 25: Aligning Structures using the Ferredoxin Sequence Alignment
25.a. Continue with the Jalview project created in exercise 23
25.b. Open the 3D Structure chooser dialog from the popup menu for FER1_SPIOL by
right-clicking its ID (CMD-click on Macs), and selecting ⇒ 3D Structure Data . . .
25.c. Pick 1A70 from the Structure Chooser dialog, and click the View button. Jalview
will give you the option of aligning the structure to the one already open. To superimpose the structure associated with FER1_MAIZE with the one associated with
FER1_SPIOL, press Yes.
The Jmol view should update to show both structures, and one will be moved on to
the other. If this doesn’t happen, use the Align function in the Jmol submenu.
25.d. Create a new view on the alignment, and hide all but columns 121 through to 132
(you can do this via View ⇒ Hide ⇒ All but selected region).
25.e. Select the newly created view in the Jmol ⇒ Superpose With submenu, and then
recompute the superposition with Jmol ⇒ Align Structures.
Note how the molecules shift position when superposed with only a small region of
the alignment.
25.f. Compare RMSDs obtained when superimposing molecules with columns 121-132 and
with the whole alignment.
25.g. The RMSD report can be viewed by right clicking the mouse on Jmol window, and
select Console from the menu (if nothing is shown, recompute the superposition after
displaying the console).
Which view do you think give the best 3D superposition, and why ?
Colouring Complexes
The ability to control which multiple alignment view is used to colour structural data is essential
when working with data relating to multidomain biomolecules and complexes.
In these situations, each chain identified in the structure may have a different evolutionary history,
and a complete picture of functional variation can only be gained by integrating data from different
alignments on the same structure view. An example of this is shown in Figure 7.5, based on data
from Song et. al.9
9 Structure of DNMT1-DNA Complex Reveals a Role for Autoinhibition in Maintenance DNA Methylation. Jikui Song,
Olga Rechkoblit, Timothy H. Bestor, and Dinshaw J. Patel. Science 2011 331 1036-1040 DOI:10.1126/science.1195380
Figure 7.4: Choosing a different view for colouring a structure display Browsing the
View ⇒ Colour by .. menu provides full control of which alignment view is used
to colour structures when the Colours ⇒ By Sequence option is selected.
Figure 7.5: The biological assembly of Mouse DNA Methyltransferase-1 coloured by
Pfam alignments for its major domains Alignments for each domain within the
Uniprot sequence DNMT1_MOUSE have been used to visualise sequence conservation in each component of this protein-DNA complex. Instructions for recreating this
figure are given in exercise 26.
Exercise 26: Colouring a Protein Complex to Explore Domain-Domain Interfaces
26.a. Download the PDB file at http://www.jalview.org/tutorial/DNMT1_MOUSE.pdb to
your desktop. This is the biological unit for PDB ID 3pt6, as identified by the PDBe’s
PISA server.
26.b. Launch the Jalview desktop and ensure you have at least 1G of free memory available.
See section 1.4.6 for how to do this or click the following link:
http: // www. jalview. org/ services/ launchApp? jvm-max-heap= 2G
26.c. Retrieve the following PFAM alignments from the PFAM (full) source : PF02008
PF01426 PF00145 (enter all three - they will each be retrieved into their own alignment window).
26.d. Drag the URL or file of the structure you downloaded in step 1 onto one of the alignments to associate it with the mouse sequence in that Pfam domain family.
26.e. Use the Find dialog to locate every DNMT1_MOUSE sequence in the alignment and
for each one, open the Structure Chooser via the ID popup menu (⇒ 3D Structure
Data . Select the DNMT1_MOUSE.pdb structure from the ‘Cached Structures’ view,
and click View.
Part of the newly opened structure will be coloured the same way as the associated
DNMT1_MOUSE sequence is in the alignment view.
WARNING: do not select all sequences and open the Structure Chooser ! This
will cause Jalview to attempt to discover all structures for sequences in the alignment.
26.f. Repeat the previous two steps for each of the other alignments. In each case, after selecting the DNMT1_MOUSE.pdb structure and hitting the ‘View’ button on the
Structure Chooser dialog, Jalview will ask if you wish to create a new Jmol view. Respond ‘Yes’ each time. This will ensure ensure each sequence fragment is associated
with the same Jmol view.
26.g. Pick a different colourscheme for each alignment, and use the Colour by .. submenu
to ensure they are all used to colour the complex shown in the Jmol window.
The different shading schemes will allow regions of strong physicochemical conservation are highlighted on the domains in the structure.
26.h. The final step needed to reproduce the shading in Figure 7.5 is to use the Colour ⇒
By Annotation. . . option in each alignment window to shade the alignment by the
Conservation annotation row (introduced in section 3.1.5).
Ensure that you first disable the View ⇒ Show Features menu option, or you may not
see any colour changes in the associated structure.
Examine the regions strongly coloured at the interfaces betweeen each protein domain, and the DNA binding region. What do you think these patterns mean ?
26.i. Save your work as a Jalview project and verify that it can be opened again by starting
another Jalview Desktop instance, and dragging the saved project into the desktop
Chapter 8
Protein sequence analysis and
structure prediction
Many of Jalview’s sequence feature and annotation capabilities were developed to allow the results
of sequence based protein structure prediction methods to be visualised and explored. This chapter
introduces services integrated with the Jalview Desktop for predicting protein secondary structure
and protein disorder.
Protein Secondary Structure Prediction
Protein secondary structure prediction is performed using the Jpred1 server at the University of
Dundee2 . The behaviour of this calculation depends on the current selection:
◦ If nothing is selected, Jalview will check the length of each alignment row to determine if the
visible sequences in the view are aligned.
- If all rows are the same length (often due to the application of the Edit ⇒ Pad Gaps
option), then a JPred prediction will be run for the first sequence in the alignment, using
the current alignment as the profile to use for prediction.
- Otherwise, just the first sequence will be submitted for a full JPred prediction.
◦ If just one sequence (or a region in one sequence) has been selected, it will be submitted to the
automatic JPred prediction server for homolog detection and prediction.
◦ If a set of sequences are selected, and they appear to be aligned using the same criteria as
above, then the alignment will be used for a JPred prediction on the first sequence in the set
(that is, the one that appears first in the alignment window).
1 “The Jpred 3 Secondary Structure Prediction Server” Cole, C., Barber, J. D. and Barton, G. J. (2008) Nucleic Acids
Research 36, (Web Server Issue) W197-W201
“Jpred: A Consensus Secondary Structure Prediction Server” Cuff, J. A., Clamp, M. E., Siddiqui, A. S., Finlay, M. and
Barton, G. J. (1998) Bioinformatics 14, 892-893
2 http://www.compbio.dundee.ac.uk/www-jpred/
Figure 8.1: Secondary Structure Prediction Status (left) and results (right) windows for
JPred predictions.
Jpred is launched in the same way as the other web services. Select Web Service ⇒ Secondary
Structure Prediction ⇒ JPred Secondary Structure Prediction3 from the alignment window menu
(Figure 8.1). A status window opens to inform you of the progress of the job. Upon completion, a new
alignment window opens and the Jpred predictions are included as annotations. Consult the Jpred
documentation for information on interpreting these results.
Hidden Columns and JPred Predictions
Hidden columns can be used to exclude parts of a sequence or profile from the input sent to the JNet
service. For instance, if a sequence is known to include a large loop insertion, hiding that section
prior to submitting the JNet prediction can produce different results. In some cases, these secondary
structure predictions can be more reliable for sequence on either side of the insertion4 . Prediction
results returned from the service will be mapped back onto the visible parts of the sequence, to
ensure a single frame of reference is maintained in your analysis.
3 JNet is the Neural Network based secondary structure prediction method that the JPred server uses.
4 This, of course, cannot be guaranteed.
Exercise 27: Secondary Structure Prediction
Note: The annotation panel can get quite busy during this exercise. Try hiding some annotations rows by right clicking the mouse in the annotation label panel and select the “Hide this
row” option. The Annotations dropdown menu on the alignment wndow also provides options
for reording and hiding autocalculated and sequence associated annotation.
27.a. Open the alignment at http://www.jalview.org/tutorial/alignment.fa. Select
the sequence FER_MESCR by clicking on the sequence ID. Then select Web Service
⇒ Secondary Structure Prediction ⇒ JPred Secondary Structure Prediction from the
alignment window menu. A status window will appear and after some time (about
2-4 min) a new window with the JPred prediction will appear. Note that the number
of sequences in the results window is many more than in the original alignment as
JPred performs a PSI-BLAST search to expand the prediction dataset. The results
from the prediction are visible in the annotation panel. JPred secondary structure
prediction annotations are examples of sequence-associated alignment annotation.
27.b. Select a different sequence and perform a JPred prediction in the same way. There
will probably be minor differences in the predictions.
27.c. Select the sequence used in the second sequence prediction by clicking on its name in
the sequence ID panel, and copy ([CTRL] or [CMD]-C) and then paste it [CTRL] or
[CMD]-V) into the first prediction window. You can now compare the two predictions
as the annotations associated with the sequence has also been copied across.
27.d. Select and hide some columns in one of the alignment profiles that were returned
from the JNet service, and then submit the profile for prediction again.
27.e. When you get the result, verify that the prediction has not been made for the hidden
parts of the profile (by clicking the mouse on column ruler and right click to open the
context menu and select Reveal All), and that the JPred reliability scores differ from
the prediction made on the full profile.
27.f. In the original alignment that you loaded in step 1, select all sequences, then open the
Sequence ID ⇒ Selection submenu by right clicking the mouse to open the context
menu, and select the Add Reference Annotation option.
All the JPred predictions for the sequences will now be visible in the original alignment window.
Homework: Go back to the last step of exercise 14 and follow the instructions to view the
Jalview annotations file created from the annotations generated by the JPred server for your
Protein Disorder Prediction
Disordered regions in proteins were classically thought to correspond to “linkers” between distinct
protein domains, but disorder can also play a role in function. The Web Service ⇒ Disorder menu in
the alignment window allows access to protein disorder prediction services provided by the configured JABAWS servers.
8.2.1 Disorder Prediction Results
Each service operates on sequences in the alignment to identify regions likely to be unstructured
or flexible, or alternately, fold to form globular domains. As a consequence, disorder predictor re-
Figure 8.2: Annotation rows for several disorder predictions on a sequence. A zoomed
out view of a prediction for a single sequence. The sequence is shaded to highlight
disordered regions (brown and grey), and the line plots below the Sequence show the
raw scores for various disorder predictors. Horizontal lines on each graph mark the
level at which disorder predictions become significant.
sults include both sequence features and sequence associated alignment annotation rows. Section
4 describes the manipulation and display of these data in detail, and Figure 8.3 demonstrates how
sequence feature shading and thresholding (described in Section 4.2.2) can be used to highlight differences in disorder prediction across aligned sequences.
Figure 8.3: Shading alignment by sequence disorder. Alignment of Interleukin IV homologs coloured with Blosum62 with protein disorder prediction sequence features
overlaid, shaded according to their score. Borderline disordered regions appear
white, reliable predictions are either Green or Brown depending on the type of disorder prediction.
Navigating Large Sets of Disorder Predictions
Figure 8.2 shows a single sequence annotated with a range of disorder predictions. Disorder prediction annotation rows are associated with a sequence in the same way as secondary structure
prediction results. When browsing an alignment containing large numbers of disorder prediction
annotation rows, clicking on the annotation row label will highlight the associated sequence in the
alignment display, and double clicking will select that sequence.
8.2.3 Disorder Predictors provided by JABAWS 2.0
For full details of each predictor and the results that Jalview can display, please consult Jalview’s
protein disorder service documentation. Short descriptions of the methods provided in JABAWS 2.0
are given below:
DisEMBL (Linding et al., 2003) is a set of machine-learning based predictors trained to recognise
disorder-related annotation found on PDB structures.
COILS Predicts loops/coils according to DSSP definitions5 . Features mark range(s) of residues predicted as loops/coils, and annotation row gives raw value for each residue. Value over 0.516 indicates
HOTLOOPS constitute a refined subset of COILS, namely those loops with a high degree of mobility
as determined from Cα temperature factors (B factors). It follows that highly dynamic loops should
be considered protein disorder. Features mark range(s) of residues predicted to be hot loops and
annotation row gives raw value for each residue. Values over 0.6 indicates hot loop.
REMARK465 “Missing coordinates in X-ray structure as defined by remark465 entries in PDB.
Nonassigned electron densities most often reflect intrinsic disorder, and have been used early on in
disorder prediction.” Features give range(s) of residues predicted as disordered, and annotation rows
gives raw value for each residue. Values over 0.1204 indicates disorder.
RONN a.k.a. Regional Order Neural Network
RONN employs an approach known as the ‘bio-basis’ method to predict regions of disorder in sequences based on their local similarity with a gold-standard set of disordered protein sequences. It
yields a set of disorder prediction scores, which are shown as sequence annotation below the alignment.
JRonn6 Annotation Row gives RONN score for each residue in the sequence. Scores above 0.5
identify regions of the protein likely to be disordered.
5 DSSP Classifications of secondary structure are: α-helix (H), 310-helix (G), β-strand (E) are ordered, and all other
states (β-bridge (B), β-turn (T), bend (S), π-helix (I), and coil (C)) considered loops or coils.
6 JRonn denotes the score for this server because JABAWS runs a Java port of RONN developed by Peter Troshin and
distributed as part of Biojava 3
IUPred employs an empirical model to estimate likely regions of disorder. There are three different
prediction types offered, each using different parameters optimized for slightly different applications. It provides raw scores based on two models for predicting regions of ‘long disorder’ and ‘short
disorder’. A third predictor identifies regions likely to form structured domains.
Long disorder Annotation rows predict context-independent global disorder that encompasses at
least 30 consecutive residues of predicted disorder. A 100 residue window is used for calculation.
Values above 0.5 indicates the residue is intrinsically disordered.
Short disorder Annotation rows predict for short, (and probably) context-dependent, disordered
regions, such as missing residues in the X-ray structure of an otherwise globular protein. Employs
a 25 residue window for calculation, and includes adjustment parameter for chain termini which
favors disorder prediction at the ends. Values above 0.5 indicate short-range disorder.
Structured domains are marked with sequence Features. These highlight likely globular domains
useful for structure genomics investigation. Post-analysis of disordered region profile to find continuous regions confidently predicted to be ordered. Neighbouring regions close to each other are merged,
while regions shorter than the minimal domain size of at least 30 residues are ignored.
GLOBPLOT defines regions of globularity or natively unstructured regions based on a running sum
of the propensity of residues to be structured or unstructured. The propensity is calculated based on
the probability of each amino acid being observed within well defined regions of secondary structure
or within regions of random coil. The initial signal is smoothed with a Savitzky-Golay filter, and
its first order derivative computed. Residues for which the first order derivative is positive are
designated as natively unstructured, whereas those with negative values are structured.
Disordered region sequence features are created marking mark range(s) of residues with positive
first order derivatives, and Globular Domain features mark long stretches of order. Dydx annotation rows give the first order derivative of smoothed score. Values above 0 indicates residue is
Smoothed Score and Raw Score annotation rows give the smoothed and raw scores used to create
the differential signal that indicates the presence of unstructured regions. These are hidden by
default, but can be shown by right-clicking on the alignment annotation panel and selecting Show
hidden annotation.
Exercise 28: Protein Disorder Prediction
Before starting this exercise, make sure you enable the ‘Add Temperature Factor’ option in
your Structures preferences.
28.a. Open the alignment at: http://www.jalview.org/tutorial/interleukin7.fa.
28.b. Run the DisEMBL disorder predictor the Service ⇒ Disorder Prediction submenu.
28.c. Select all the sequences, and open the Structure Chooser via the Sequence ID ⇒ 3D
Structure Data. . . popup menu. Hit the View button to retrieve and show all PDB
structures for the sequences.
28.d. Compare the disorder predictions to the structure data by mapping any available
temperature factors to the alignment via the Sequence ID Popup ⇒ Selection ⇒ Add
reference annotation option.
28.e. Apply the IUPred disorder prediction method. Use the Per sequence option in the
Colour ⇒ By annotation . . . dialog to shade the sequences by the long and short
disorder predictors. Note how well the disordered regions predicted by each method
agree with the structure.
Chapter 9
DNA and RNA Sequences
Working with DNA
Jalview was originally developed for the analysis of protein sequences, but now includes some specific features for working with nucleic acid sequences and alignments. Jalview recognises nucleotide
sequences and alignments based on the presence of nucleotide symbols [ACGT] in greater than 85%
of the sequences. Built in codon-translation tables can be used to translate ORFs into peptides for
further analysis. ENA nucleotide records retrieved via the sequence fetcher (see Section 1.4.5) are
also parsed in order to identify codon regions and extract peptide products. Furthermore, Jalview
records mappings between protein sequences that are derived from regions of a nucleotide sequence.
Mappings are used to transfer annotation between nucleic acid and protein sequences, and to dynamically highlight regions in one sequence that correspond to the position of the mouse pointer in
9.1.1 Alignment and Colouring
Jalview provides a simple colourscheme for DNA bases, but does not apply any specific conservation
or substitution score model for the shading of nucleotide alignments. However, pairwise alignments
performed using the Calculate ⇒ Pairwise Alignment . . . option will utilise an identity score matrix
to calculate alignment score when aligning two nucleotide sequences.
Aligning Nucleic Acid Sequences
Jalview has limited knowledge of the capabilities of the programs that are made available to it via
web services, so it is up to you, the user, to decide which service to use when working with nucleic
acid sequences. The table shows which alignment programs are most appropriate for nucleotide
alignment. Generally, all will work, but some may be more suited to your purposes than others.
We also note that none of these include support for taking RNA secondary structure prediction into
account when aligning sequences (but will be providing services for this in the future!)
NA support
Yes (treat U as T)
Default is to autodetect nucleotide sequences.
Editable parameters include nucleotide substitution matrices and distance metrics.
Default is to autodetect nucleotide sequences.
Editable parameters include nucleotide substitution matrices and distance metrics.
Will autodetect nucleotide sequences and use
a hardwired substitution model (all aminoacid sequence related parameters are ignored). Unknown whether substitution model
treats Uracil specially.
ProbCons has no special support for aligning
nucleotide sequences. Whilst an alignment
will be returned, it is unlikely to be reliable.
Sequence type is automatically detected and
an appropriate parameter set used as required. A range of nucleotide specific score
models are available.
Table 9.1: JABAWS Alignment programs suitable for aligning nucleic acid sequences.
All JABAWS alignment services will return an alignment if provided with RNA or
DNA sequences, with varying reliability.
9.1.2 Translate cDNA
The Calculate ⇒ Translate cDNA function in the alignment window is only available when working
with a nucleic acid alignment. It uses the standard codon translation table given in the online help
documentation to translate a nucleotide alignment, or the currently selected region, into a set of
aligned peptide sequences. Any features or annotation present on the nucleotide alignment will also
be translated, allowing DNA alignment analysis results to be transferred on to peptide products for
further investigation.
9.1.3 Linked DNA and Protein Views
Views of alignments involving DNA sequences are
linked to views of alignments containing their peptide products in a similar way to views of protein sequences and views of their associated structures. Peptides translated from cDNA that have been fetched
from ENA records for DNA contigs are linked to their
‘parent’ coding regions. Mousing over a region of the
peptide highlights codons in views showing the original coding region.
Coding Regions from ENA Records
Many ENA records that can be retrieved with the sequence fetcher contain exons. Coding regions will
be marked as features on the ENA nucleotide sequence, and Uniprot database cross references will be
listed in the tooltip displayed when the mouse hovers over the sequence ID. Uniprot database cross
references extracted from ENA records are sequence cross references, and associate a Uniprot sequence’s coordinate system with the coding regions annotated on the ENA sequence. Jalview utilises
cross-reference information in two ways.
Retrieval of Protein or DNA Cross References
The Calculate ⇒ Get Cross References function is only available when Jalview recognises that there
are protein/DNA cross-references present on sequences in the alignment. When selected, it retrieves
the cross references from the alignment’s dataset (a set of sequence and annotation metadata shared
between alignments) or using the sequence database fetcher. This function can be used for ENA
sequences containing coding regions to open the Uniprot protein products in a new alignment window. The new alignment window that is opened to show the protein products will also allow dynamic
highlighting of codon positions in the ENA record for each residue in the protein product(s).
Retrieval of Protein Features on Coding Regions
The Uniprot cross-references derived from ENA records can be used by Jalview to visualize protein
sequence features directly on nucleotide alignments. This is because the database cross references
include the sequence coordinate mapping information to correspond regions on the protein sequence
with that of the nucleotide contig. Jalview will use the Uniprot accession numbers associated with
the sequence to retrieve features, and then map them onto the nucleotide sequence’s coordinate
system using the coding region location.
Exercise 29: Visualizing Protein Features on Coding Regions
29.a. Use the sequence fetcher to retrieve ENA record D49489.
29.b. Ensure that View ⇒ Show Sequence Features is checked and change the alignment
view format to Wrapped mode so the distinct exons can be seen.
29.c. Open the DAS Settings tab in the Sequence Feature Settings. . . window View ⇒
Features setting and fetch features for D49489 from the Uniprot reference server,
and any additional servers that work with the Uniprot coordinate system.
29.d. Mouse over the features retrieved, note that they have been mapped onto the coding
regions, and in some cases broken into several parts to cover the distinct exons.
29.e. Open a new alignment view containing the Uniprot protein product with Calculate ⇒
Get Cross References ⇒ Uniprot and examine the database references and sequence
features. Experiment with the interactive highlighting of codon position for each
Figure 9.1: Uniprot and PDB sum features retrieved and mapped onto coding regions of ENA
record V00488 (an earlier version of Jalview is shown here).
Working with RNA
Jalview allows the creation of RNA secondary structure annotation, and includes the VARNA secondary structure viewer for the display of RNA base pair diagrams. It also allows the extraction of
RNA secondary structure from 3D data when available.
Performing RNA Secondary Structure Predictions
Secondary structure consensus calculations can be performed by enabling the VIENNA service via
the Web Service ⇒ Secondary Structure menu. These consensus structures are created by analysing
the covariation patterns in all visible sequences on the alignment. For more information see the
VIENNA documentation.
Figure 9.2: Secondary structure consensus calculations can be performed by enabling the VIENNA service via the Web Service ⇒ Secondary Structure menu.
Figure 9.3: VIENNA can calculate alternate RNA base pairing probabilities. These are shown
in Jalview as tool-tips on the RNA secondary structure probability score.
Exercise 30: Viewing RNA Structures
30.a. Import RF00162 from the Rfam (Seed) source using Fetch sequence(s) from the Desktop’s File menu.
30.b. Select Colour by RNA Helices to shade the alignment by the secondary structure
annotation provided by Rfam.
30.c. Open VARNA with Structure ⇒ View Structure ⇒ RNA Secondary Structure. In
the VARNA Structures Manager toggle between (i) secondary structure (alignment)
(with gaps) and (ii) trimmed secondary structure (alignment). Explore the difference
between trimmed and untrimmed views. Click on different residues in the VARNA
diagram - you should also see them highlighted and selected in the sequence alignment window.
30.d. In the VARNA Structures Manager, right click on display window to bring up the
pop up context menu. Explore the options within the File, Export, Display and Edit
VARNA views are stored in Jalview project files, in the same way as 3D structure views
produced by Jmol and Chimera.
30.e. Enable the calculation and display of an RNAAliFold secondary structure prediction for the alignment by selecting Web Service ⇒ Secondary Structure Prediction ⇒
RNAAliFold .
30.f. Edit the RNAAliFold calculation settings to show Base Pair probabilities. Explore
how editing the alignment affects the consensus calculation.
30.g. Import 2GIS from the PDB database into a new window with Fetch sequence(s).
30.h. Click on a sequence in Sequence ID panel and select Structure ⇒ View Structure ⇒
2GIS, to view the structure in Jmol window. Click on different residues and located
them in the sequence alignment window.
Chapter 10
The term “Webservices” refers to a variety of data exchange mechanisms based on HTTP.1
Jalview can exploit public webservices to access databases remotely,
and also submit data to public services by opening pages with your
web browser. These types of services are ‘one-way’, i.e. data is either
sent to the webservice or retrieved from it by Jalview. The desktop application can also interact with ‘two-way’ remote analysis services in
order to offload computationally intensive tasks to High Performance
Computing facilities. Most of these two-way services are provided by
Java Bioinformatics Analysis Web Service (JABAWS) servers2 , which
provides an easily installable system for performing a range of bioinformatics analysis tasks.
One-Way Web Services
There are two types of one way service in Jalview. Database services, which were introduced in in
Section 1.4.5, provide sequence and alignment data. They can also be used to add sequence IDs to
an alignment imported from a local file, prior to further annotation retrieval, as described in Section
Remote Analysis Web Services
Remote analysis services enable Jalview to use external computational facilities. There are currently
three types of service - multiple sequence alignment, protein secondary structure prediction, and
alignment analysis. Many of these are provided by JABA servers, which are described at the end of
this section. In all cases, Jalview will construct a job based on the alignment or currently selected
sequences, ask the remote server to run the job, monitor status of the job and, finally, retrieve the
results of the job and display them. The Jalview user is kept informed of the progress of the job
1 HTTP: Hyper-Text Transfer Protocol.
2 See http://www.compbio.dundee.ac.uk/jabaws for more information and to download your own server.
through a status window.
Currently, web service jobs and their status windows are not stored in Jalview Project Files3 , so
it is important that you do not close Jalview whilst a job is running. It is also essential that you
have a continuous network connection in order to successfully use web services from Jalview, since
it periodically checks the progress of running jobs.
JABA Web Services for Sequence Alignment and Analysis
JABA stands for “JAva Bioinformatics Analysis”, which is a system developed by Peter Troshin and
Geoff Barton at the University of Dundee for running computationally intensive bioinformatics analysis programs. A JABA installation typically provides a range of JABA web services (JABAWS) for
use by other programs, such as Jalview.
Exercises in the remainder of this section will demonstrate the simplest way of installing JABA on
your computer, and configuring Jalview so it can access the JABA services. If you need any further
help or more information about the services, please go to the JABAWS home page.
Changing the Web Services Menu Layout
If you are working with a lot of different JABA services, you may wish to change the way Jalview
lays out the web services menu. You can do this from the Web Services tab of the Preferences dialog
Exercise 31: Changing the Layout of the Web Services Menu
31.a. Make sure you have loaded an alignment into Jalview, and examine the current layout of the alignment window’s Web Service menu.
31.b. Open the preferences dialog box and select the web services tab.
31.c. Ensure the Enable JABAWS services checkbox is selected, and unselect the Enable
Enfin Services checkboxes.
31.d. Hit Refresh Services to update the web services menu – once the progress bar has
completed, open the Web Service menu to view the changes.
31.e. Select the Index by host checkbox and refresh the services once again.
Observe the way the layout of the JABAWS Alignment submenu changes.
31.f. Do the same with the Index by type checkbox.
Jalview provides these options for configuring the layout of the Web Service menu because different
Jalview users may have access to a different number of JABA services, and each will have their own
preference regarding the layout of the menu.
3 This may be rectified in future versions.
Figure 10.1: The Jalview Web Services preferences panel. Options are provided for configuring the list of JABA servers that Jalview will use, enabling and disabling Enfin
services, and configuring the layout of the web services menu.
Testing JABA services
The JABAWS configuration dialog shown in Figure 10.1 has colour codes to indicate whether the
Desktop can access the server, and whether all services advertised by the server are functional. The
colour codes are:
• Red - Server cannot be contacted or reports a connection error.
• Amber - Jalview can connect, but one or more services are non-functional.
• Green - Server is functioning normally.
Test results from JABAWS are reported on Jalview’s console output (opened from the Tools menu).
Tests are re-run every time Jalview starts, and when the [Refresh Services] button is pressed on the
Jalview JABAWS configuration panel.
Resetting the JABA Services Setting to their Defaults
Once you have configured a JABAWS server and selected the OK button of the preferences menu,
the settings will be stored in your Jalview preferences file, along with any preferences regarding
the layout of the web services menu. If you should ever need to reset the JABAWS server list to its
defaults, use the ‘Reset Services’ button on the Web Services preferences panel.
Running your own JABA Server
You can download and run JABA on your own machine using the ‘VMWare’ or VirtualBox virtual
machine environments. If you would like to do this, there are full instructions at the JABA web site.
Exercise 32: Installing a JABA Virtual Machine on your Computer
This tutorial will demonstrate the simplest way of installing JABA on your computer, and
configuring Jalview so it can access the JABA services.
You will need a copy of VMWare Player/Workstation/Fusion on your machine.
32.a. If you do not have VMWare player installed, download it from www.vmware.com (this
takes a few minutes – you will need to register and wait for an email with a download
32.b. Download the JABA virtual appliance archive called ‘jaba-vm.zip’ from
WARNING: This is large (about 300MB) and will take some time to download.
32.c. Unpack the archive’s contents to a place on your machine with at least 2GB of free
space (On Windows, right click on the archive, and use the ’Extract archive..’ option).
32.d. Open the newly extracted directory and double click the VMWare virtual machine
configuration file (jabaws.vcf). This will launch the VMWare player.
32.e. Once VMWare player has started up, it may ask the question “Did you move or copy
this virtual appliance?” – select ‘Copy’.
32.f. You may be prompted to download the VMWare linux tools. These are not necessary,
so close the window or click on ‘Later’.
32.g. You may also be prompted to install support for one or more devices (USB or otherwise). Say ‘No’ to these options.
32.h. Once the machine has loaded, it will display a series of IP addresses for the different
services provided by the VM. Make a note of the JABAWS URL – this will begin with
‘http:’ and end with ‘/jabaws”.
Exercise 33: Configuring Jalview to Access your new JABAWS Virtual Appliance
33.a. Start Jalview (If you have not done so already).
33.b. Enable the Jalview Java Console by selecting its option from the Tools menu.
Alternately, use the System Java console if you have configured it to open when
Jalview is launched, via your system’s Java preferences (under the ‘Advanced’ tab
on Windows).
33.c. Open the Preferences dialog and locate the Web Services tab.
33.d. Add the URL for the new JABAWS server you started in Exercise 32 to the list of
JABAWS urls using the ‘New Service URL’ button.
33.e. You will be asked if you want to test the service. Hit ‘Yes’ to do this – you should then
see some output in the console window.
Take a close look at the output in the console. What do you think is happening?
33.f. Hit OK to save your preferences – you have now added a new JABA service to Jalview!
33.g. Try out your new JABA services by loading the ferredoxin sequences from
33.h. Launch an alignment using one of the JABA methods provided by your server. It will
be listed under the JABAWS Alignment submenu of the Web Service menu on the
alignment window.
Note: You can watch the JABA VM appliance’s process working by opening the process monitor on your system. (On Windows XP, this involves right-clicking the system
clock and opening the task manager – then selecting the ’Processes’ tab and sort by
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF