BRIG 0.

95 Manual
Nabil Alikhan

June 27, 2011

http://sourceforge.net/projects/brig/

1
CONTENTS 2

Contents
1 Introduction 3

2 Licence 9

3 Installation 9
3.1 Installing BLAST . . . . . . . . . . . . . . . . . . . . . . . . . . 9

4 Warning when using BLAST 11

4.1 Low complexity filtering . . . . . . . . . . . . . . . . . . . . . . 11
4.2 Expected values(e-values) and bit scores . . . . . . . . . . . . . . 11

5 Visualising whole genome comparisons 13

5.1 Step 1: Load in sequences . . . . . . . . . . . . . . . . . . . . . 13
5.2 Step 2: Configure rings . . . . . . . . . . . . . . . . . . . . . . . 14
5.3 Step 3: Review and submit . . . . . . . . . . . . . . . . . . . . . 16

6 Working with a Multi-FASTA reference 18

6.1 Step 1: Load in sequences . . . . . . . . . . . . . . . . . . . . . 18
6.2 Step 2: Configure rings, annotations and spacer value . . . . . . . 19
6.3 Step 3: Configure image settings and submit . . . . . . . . . . . . 23

7 Visualising graphs and genome assemblies 25

7.1 Walkthrough for visualising SAM file mapping coverage. . . . . . 26
7.2 Walk through for visualising ace file assembly coverage. . . . . . 31

8 Walkthroughs on creating custom annotations 39

8.1 Adding custom annotations from a tab-delimited file, GenBank or
EMBL file . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
8.1.1 Step 1: Load in sequences . . . . . . . . . . . . . . . . . 39
8.1.2 Step 2: Configure rings . . . . . . . . . . . . . . . . . . . 40
8.1.3 Step 3: Adding annotations . . . . . . . . . . . . . . . . . 41
8.1.4 Step 4: Review and submit . . . . . . . . . . . . . . . . . 43
8.2 How to create tab-delimited files for BRIG . . . . . . . . . . . . . 45

9 Configuration options 46
9.1 Saving and reopening your work . . . . . . . . . . . . . . . . . . 46
9.2 BLAST options . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
9.3 Setting BRIG options . . . . . . . . . . . . . . . . . . . . . . . . 48
9.4 Setting Image options . . . . . . . . . . . . . . . . . . . . . . . . 50
9.5 Loading a preset image template . . . . . . . . . . . . . . . . . . 52
1 INTRODUCTION 3

1 Introduction
The BLAST Ring Image Generator (BRIG) is a cross-platform desktop applica-
tion written in Java 1.6. It uses CGView[5] for image rendering and the Basic
Local Alignment Search Tool (BLAST) for genome comparisons. It has a graph-
ical user interface programmed on the Swing framework, which takes the user
step-by-step through the configuration of a circular image generation. Figure 1 is
an example of an image BRIG can create.

Figure 1: BRIG example output image of a simulated draft E. coli O157:H7

genome. The figure show BLAST comparisons against 28 published E. coli and
Salmonella genomes against the simulated draft genome.
1 INTRODUCTION 4

Figure 2 shows a magnified view of the same example image showing similar-
ity between a central reference genome in the centre against other query sequences
as a set of concentric rings, where colour indicates a BLAST match of a particular
percentage identity. BRIG does not represent sequences that are not present in the
reference genome The image shows:

• GC skew,

• GC content,

• Genome coverage and contig boundaries (calculated from an assembly file),

• Genome alignment results, customs annotations.

Figure 2: A magnified view of BRIG example image

1 INTRODUCTION 5

How to use this manual

This manual contains a set of detailed walk throughs where readers are taken
step by step through a worked example. Each walkthrough highlights different
features of BRIG and users should work through each one. If you are interested
in a particular aspect of BRIG, please turn to the relevant walkthrough:

• Whole genome comparisons, including how to load in coverage graphs, e.g.

Figures 1 & 3, see Section 5 on page 13.

• Using a user-defined list of genes as a reference (in Multi-FASTA), e.g Fig-

ure 4, see Section 6 on page 18.

• Creating and visualising graphs generated from assembly (.ace) or read

mapping coverage (.SAM), e.g Figure 5, see Section 7 on page 25.

• Labeling images with information from GenBank, Tab-delimited or Multi-

FASTA files, like those seen in Figure 3, 4 & 5, see Section 8 on page
39.

The manual also has detailed instructions for how to install and configure BRIG:

• For instructions on how to install BRIG, see Section 3 on page 9.

• For instructions on how to configure BRIG and save BRIG settings, see
Section 9 on 46.
1 INTRODUCTION 6

Figure 3: Reference: Published E.coli O157:H7 Sakai genome. Query: Com-

plete genome sequences of related strains, listed in the key. The prophage regions
from the Sakai genome are marked in alternating black & blue. To make an image
like this please refer to Section 5 on page 13.
1 INTRODUCTION 7

Figure 4: Reference: A list of translated genes that make up the Locus of Entero-
cyte Effacement (LEE), which encodes a Type III secretion system. Query: Raw
sequencing reads simulated from several complete LEE+ published genomes (nu-
cleotide sequence) and E. coli K12, (negative control; LEE-). You can clearly see
gene presence/absence, and divergence (the colour represents sequence identity
on a sliding scale, the greyer it gets; the lower the percentage identity). To make
an image like this please refer to Section 6 on page 18.
1 INTRODUCTION 8

Figure 5: Reference: Published E.coli O157:H7 Sakai genome. Query: Read

mapping coverage of sequencing reads simulated from complete genomes, indi-
cated in the key. Simulated sequencing reads were mapped onto the published
complete Sakai genome using BWA. The read coverage for each genome was
generated from the resulting SAM files. Compare this with Figure 3, which is
based on the original published genome sequences. To make an image like this
please refer to Section 7 on page 25.
2 LICENCE 9

2 Licence
This program is free software: you can redistribute it and/or modify it under the
terms of the GNU General Public License as published by the Free Software Foun-
dation, either version 3 of the License, or (at your option) any later version.
This program is distributed in the hope that it will be useful, but without any
warranty; without even the implied warranty of merchantability or fitness for a
particular purpose. See the GNU General Public License for more details.
You should have received a copy of the GNU General Public License along
with this program. If not, see <http://www.gnu.org/licenses/>.
Please note that these restrictions do not apply to the third party libraries
bundled with this software.

3 Installation
There’s no real ”Installation” process for BRIG itself. However, BLAST+[2] or
BLAST legacy[1] must already be installed and BRIG needs to be able to locate
the BLAST executables (See Section 3.1).
To run BRIG users need to:

1. Download the latest version (BRIG-x.xx-dist.zip) from

http://sourceforge.net/projects/brig/

2. Unzip BRIG-x.xx-dist.zip to a desired location.

3. Run BRIG.jar, by double clicking.

Users who wish to run BRIG from the command-line need to:

1. Navigate to the unpacked BRIG folder in a command-line interface (termi-

nal, console, command prompt).

2. Run “java -Xmx1500M -jar BRIG.jar”. Where -Xmx specifies the amount
of memory allocated to BRIG.

3.1 Installing BLAST

The latest version of BLAST+[2] can be downloaded from:
ftp://ftp.ncbi.nlm.nih.gov/blast/executables/blast+/LATEST/
BLAST+ offers a number of improvements on the original BLAST implementa-
tion and comes as a bundled installer, which will walk users through the installa-
tion process. Please read the published paper on BLAST+:
3 INSTALLATION 10

Camacho, C., G. Coulouris, et al. (2009).“BLAST+: architecture and appli-

cations.” BMC Bioinformatics 10(1): 421 Available online at : http://www.
biomedcentral.com/1471-2105/10/421

The latest version of BLAST legacy[1] can be downloaded from:

ftp://ftp.ncbi.nlm.nih.gov/blast/executables/release/LATEST/

BLAST legacy comes as a compressed package, which will unzip the BLAST
binaries where ever the package is. We advise users to first create a BLAST direc-
tory (in either the home or applications directory), copy the downloaded BLAST
package to that directory and unzip the package.
BRIG supports both BLAST+ & BLAST Legacy. Users can specify the loca-
tion of their BLAST installation in the BRIG options menu which is:
Main window >Preferences >BRIG options.
The window is shown in Figure 6. If BRIG cannot find BLAST it will prompt
users at runtime.

PRO TIP 1: BRIG uses BLAST, do not use wwwblast or netblast with BRIG.
PRO TIP 2: If BOTH BLAST+ and legacy versions are in the same location,
BRIG will prefer BLAST+.

Figure 6: You can change where BRIG looks for BLAST in the BRIG options
window. For more information about BRIG options see Section 9.2 on page 48.
4 WARNING WHEN USING BLAST 11

4 Warning when using BLAST

BRIG relies on the Basic Local Alignment Search Tool (BLAST) for genome
comparisons. BLAST has a number of behaviours that may seem counterintuitive
and we encourage users to learn about local alignment and the BLAST algorithm
to fully understand the images that BRIG produces. There are a few concepts to
keep in mind when using BRIG:

4.1 Low complexity filtering

PRO TIP 3: BLAST filters may cause gaps in alignments, which will show up
as blank regions in BRIG images.

BLAST filters (BLAST legacy -F flag or BLAST+ -dust/-seg no flag) filter the
query sequence for low-complexity sequences by default. This includes sequences
that are highly repetitive or contain the same nucleotide for long lengths of the
sequences. Low-complexity filtering is generally a good idea, but it may break
long matches into several smaller matches.
This is often shown in BRIG images as truncations or gaps in alignments, it
is particularly obvious in very small reference sequences where alignments are
shown on a gene-by-gene level.
To prevent this, either turn off filtering or use soft masking.

4.2 Expected values(e-values) and bit scores

PRO TIP 4: BLAST’s bitscore filtering may cause different results in BRIG if
users swap the query and reference sequences, particularly if these are very
different sizes.

BLAST uses statistical thresholds to filter out “bad alignments”; alignment matches
that appear random to BLAST. One of these thresholds is the e-value, which is the
probability of the alignment occurring by chance, given the complexity of the
match, sequence composition and the size of the database. It is more likely in
a larger sequence that an alignment could occur by chance, so BLAST is more
critical of these matches.
This can create different expected values if BLAST is used with the same
reference sequence against databases of different sizes and may potentially filter
out significant matches or include poor scoring ones.
4 WARNING WHEN USING BLAST 12

Because of this, users might notice different results in BRIG images if they
swap the order of the database and reference sequences around in the BLAST,
especially if the two sequences are quite different in size. The differences are
often due to a few very low-scoring hits.
Users should consider what an appropriate e-value threshold is for the compar-
isons that they run. Remember, that BLAST runs with an e-value of 10 by default,
we recommend that users change this value. Users can set the final threshold
(e-value) with the -e flag in BLAST legacy or -evalue flag in BLAST+.

PRO TIP 5: BLAST does not handle spaces in filenames, BRIG will prompt
users if they have spaces in file locations.
5 VISUALISING WHOLE GENOME COMPARISONS 13

5 Visualising whole genome comparisons

In this section we will walk through the basics of generating an image. This walk
through will be comparing an E. coli genome with five other E. coli genomes and
mapping the read coverage from the underlying genome assembly onto the same
image. For this walk through, users will need BRIG examples.zip, which is avail-
able from the BRIG website (http://sourceforge.net/projects/brig/
files/). This contains all the genomes and files needed to follow along with this
walk through. Unzip it somewhere easily accessible, like the home directory or
desktop.

About the reference genome

The reference genome used in this walk through is a simulated E. coli genome
assembly. We took the published E. coli O157:H7 Sakai genome (Accession num-
ber BA000007) sequence and had assembly reads simulated by METASIM[4] and
then assembled these using Newbler version 2.3. The resulting contiguous se-
quences were ordered using Mauve[3] against the published Sakai genome. This
simulated E. coli is useful for illustrating some of BRIG’s graphing features for
assembly read coverage.
Enterohemorrhagic E. coli are gram-negative, enteric bacterial pathogens. They
can cause diarrhea, hemorrhagic colitis, and hemolytic uremic syndrome. This
particular genome we are using in this example was based on an E. coli O157:H7
isolated from the Sakai, Japan outbreak.

5.1 Step 1: Load in sequences

The walk through will work out of the unzipped BRIG examples.zip in the Chap-
ter5 6 8 wholeGenomeExamples folder. The walk through and related figures
will use C:\BRIG examples\Chapter5 6 8 wholeGenomeExamples as that loca-
tion.
To keep the final image consistent with the walk through, please open ”Exam-
pleProfile.xml” from the Chapter5 6 8 wholeGenomeExamples folder. This file
configures BRIG to the same image settings in the walk through.

1. First, set BRIGExample.fna as the reference sequence.

2. Set <unzipped BRIG examples folder>\Chapter5 6 8 wholeGenomeExamples

as the query sequence folder.

3. Press “add to data pool”, this should load several items into the pool list,
there should be nine files.
5 VISUALISING WHOLE GENOME COMPARISONS 14

4. Set the Chapter5 6 8 wholeGenomeExamples as the output folder.

5. The BLAST options box should be left blank.

6. Click next

PRO TIP 6: Users can add individual files to the data pool too.

5.2 Step 2: Configure rings

The next step is to configure what information is shown on each of the concentric
rings in BRIG. Create six rings, for each ring:

1. Set the legend text for each ring

2. Select the required sequences from the data pool and click on “add data” to
add to the ring list.

3. Choose a colour

4. Set the upper (90) and lower (70) identity threshold.

5 VISUALISING WHOLE GENOME COMPARISONS 15

5. Click on “add new ring” and repeat steps for each new ring required.

The values required for each ring are detailed in the table below. Notice that that

sequences can be collated into a single ring, like the example of K12 & HS. The
ring will show BLAST matches from both HS and K12.
Legend text Required sequences Colour
GC Content GC Content Ignore
GC Skew GC Skew Ignore
Coverage BRIGExample.graph 153,0,0
O157:H7 E coli O157H7Sakai.gbk 0,0,153
HS and K12 E coli HS.fna 0,153,0
E coli K12MG1655.fna
CFT073 and UTI89 E coli CFT073.fna 153,0,153
E coli UTI89.fna
5 VISUALISING WHOLE GENOME COMPARISONS 16

PRO TIP 7: Rings can be reordered by dragging them in the Ring List pane.
PRO TIP 8: You can set default threshold values in “BRIG options”. See
section 9.2 (page 48) for more details.
PRO TIP 9: When using a Genbank/EMBL file as a reference, users can
choose whether to use the protein or nucleotide sequence.

5.3 Step 3: Review and submit

The last window allows us to change the BLAST options, the location of the
image file and set the image title, which will appear in the centre of the ring. For
the walkthrough configure the third window as:

1. Set the image title as “BRIG example image”.

2. Hit submit.

3. The image will be created in the specified output directory and should look
something like Figure 7.

BRIG will format Genbank files, run BLAST, parse the results and render the im-
age. The final image (Figure 7) shows GC Content and Skew, the Genome cover-
age, contig boundaries, and the BLAST results against the other E. coli genomes.
The results for HS and K12 have been collated into a single ring, likewise for
UTI89 and CFT073.
5 VISUALISING WHOLE GENOME COMPARISONS 17

Figure 7: The final BRIG image

PRO TIP 10: Image settings, like size, fonts, etc can be configured in: Main
window >Preferences >Image options..
6 WORKING WITH A MULTI-FASTA REFERENCE 18

6 Working with a Multi-FASTA reference

6.1 Step 1: Load in sequences
This section is a walk through of how to use BRIG to generate an image using a list
of genes in Multi-FASTA format as a reference. The multi-FASTA file in this ex-
ample is a number of virulence genes from enterohemorrhagic and uropathogenic
E. coli, which includes EHEC polar fimbraie (ecpA to ecpR), EHEC Locus of En-
terocyte Effacement (espF to espG) and the UPEC F1C Fimbraie (focA to focI),
which will be compared against the whole genome seqeuences of E. coli strains
O157:H7 Sakai, K12 MG1665, O126:H7 and CFT073. Start a new session in
BRIG and load in the files from the Chapter5 6 8 wholeGenomeExamples folder
in the unzipped BRIG-Example folder:

1. Set the reference sequence as “Ecoli vir.fna”. Users can use the browse
button to traverse the file system.

2. Set <unzipped BRIG examples folder>/Chapter5 6 8 wholeGenomeExamples

as the query sequence folder.

3. Press “add to data pool”, this should load several items into the pool list.

4. Set the output folder as unzipped BRIG-Example folder.

5. Make sure the BLAST options box is blank.

6. Click “Next”.
6 WORKING WITH A MULTI-FASTA REFERENCE 19

6.2 Step 2: Configure rings, annotations and spacer value

The next step is to configure what information is shown on each concentric ring in
BRIG. Figure 8 is an example of how one of the windows should be set up. There
should be five rings. Do the follow for each ring, according to the table below:
1. Set legend text for each ring.
2. Select the required sequences from the data pool and click “add data” to
add.
3. Choose a colour
4. Set upper (90) and lower (70) identity thresholds.
5. Click “add new ring” and repeat these steps for each new ring.
Legend text Required sequences Colour
O157:H7 E coli O157H7Sakai.gbk 172,14,225
O126:H7 E coli O126.fna 255,0,51
CFT073 E coli CFT073.fna 0,0,102
K12 E coli K12MG1655.fna 161,221,231
null none ignore
After each ring is configured, users need to make the following changes:
1. Set the spacer field to 50 base pairs.
2. Set the ring size of ring 5 as “2”.

PRO TIP 11: The Spacer field determines the number of base pairs to leave
between FASTA sequences.

The next step is to add the gene annotations, which will be fetched from the Multi-
FASTA headers:
1. Click Add custom features in the second BRIG window to bring up custom
annotation window (Figure 9).
2. Double click “Ring 5”.
3. Set “input data” as Multi-FASTA.
4. Set “colour” as alternating red-blue
5. Click add.
6 WORKING WITH A MULTI-FASTA REFERENCE 20

Figure 8: Ring set-up for Multi-FASTA file

6 WORKING WITH A MULTI-FASTA REFERENCE 21

Figure 9: Custom annotation window - adding gene annotations

6 WORKING WITH A MULTI-FASTA REFERENCE 22

This step colours the gaps between FASTA entries, the gaps are calculated
from the Multi-FASTA file (Figure 10). For each genome ring, do the following:

1. Set “input data” as Multi-FASTA.

2. Set “colour” as black

3. Check “load gaps only”.

4. Click add.

The results should be similar to Figure 10 in the left hand pane. Close the window
when this is done.

PRO TIP 12: A spacer value can be set when using protein sequences from a
Genbank/EMBL file.

Figure 10: Custom annotation window - adding spacers

6 WORKING WITH A MULTI-FASTA REFERENCE 23

6.3 Step 3: Configure image settings and submit

There are a few more steps to complete and then the image is finished. In the
customize ring window:

1. Make the following changes in Preferences >Image options

(a) Set “show shading” in “Global settings” as false.

(b) Set “featureSlot” spacing in “Feature settings” as x-small.

2. Return to customize ring window, click “Next” to go to the final BRIG

confirmation window

3. Set the image title as “Various E. coli virulence genes” and press submit.

The output image should be something like Figure 11.The alternating red-blue op-
tion has automatically alternated the red and blue colours for the gene labels. This
option is available whenever a multi-FASTA file is used as a reference sequence.
This same option could be used to show contig or genome scaffold boundaries.
This image shows some real biological information very clearly.

1. CFT073 (UPEC) and K12 MG1655 (Commensal) do not carry the Locus of
Enterocyte Effacement. These virulence factors are specific to EHEC and
EPEC.

2. All E. coli shown carry the common pilus (ecpA-R).

3. Only CFT073 carries the F1C fimbriae.

PRO TIP 13: You use protein sequences as a multi-FASTA reference and use
blastx to improve alignment accuracy for divergent sequences.
6 WORKING WITH A MULTI-FASTA REFERENCE 24

Figure 11: Output image from Multi-FASTA walkthrough. This was generated
using BLAST+[2], BLAST legacy[1] will produce slightly different results.
7 VISUALISING GRAPHS AND GENOME ASSEMBLIES 25

7 Visualising graphs and genome assemblies

BRIG can produce any user-specified graph e.g coverage, read mapping, expres-
sion data etc. For example, the coverage graph in Figure 2 was produced from a
tab-delimited text file, with a start, stop and value for that range.
BRIG supports .ace files (produced by Newbler, 454/Roche’s propriety as-
sembler, and used by PHRAP/Consed) or SAM files (used for read mapping and
some de-novo assemblers). BRIG has a number of modules for handling assembly
information. These tools are:
• Contig mapping: BRIG will use BLAST to try and map contigs from an
.ace or Multi-FASTA file onto a reference genome and produce a .graph file
that can show frequency of BLAST hits and the best BLAST hit position of
contigs. It will then produce a .graph file of the frequency of BLAST hits
and the best BLAST hit position of contigs and another .graph file, with the
suffix ”rep.graph” showing all the other BLAST hits.
• Coverage graph: BRIG requires an .ace or .sam file and an output loca-
tion (Figure 7.1). BRIG will calculate coverage values over a user-defined
window and produce a .graph file in the output folder. This will create a
tab-delimited .graph file, which can be loaded into back BRIG.
• Convert graph: A draft genome is usually modified post assembly; adding
spacers, reordering contigs, etc. These changes are often not reflected in the
original .ace files BRIG uses to generate coverage graphs. BRIG can use
BLAST to align the original assembly output with the newer sequence and
map the coverage information to the new sequence. BRIG requires:
– Original 454AllContigs.fna produced by Newbler.
– Graph file created by BRIG’s “Coverage graph” module, based on
Newbler’s ace file.
– The modified sequence or another suitable reference genome.
BRIG will produce a new .graph file in the output folder, using the filename
of the original file, with “new.graph” appended to the end.
To create work with graph files: Main window >Modules >Create graph files

Using graph files in BRIG images

.graph files should visible when users load a directory into the query sequence
pool (Figure 7.1). Graphs can be treated like any other sequence file in BRIG; the
example from Figure 7.1 shows a graph file loaded into the first ring of a particular
BRIG session.
7 VISUALISING GRAPHS AND GENOME ASSEMBLIES 26

PRO TIP 14: Graph files cannot be shown on same ring as sequence files
(protein or nucleotide).

7.1 Walkthrough for visualising SAM file mapping coverage.

This section will give a worked example of producing a BRIG image show-
ing mapping reads coverage from a SAM input file. The final image will look
like Figure 12. This walkthrough requires BRIG examples.zip from the BRIG
website (http://sourceforge.net/projects/brig/files/). Unzip this
somewhere convenient. The general procedure is to first generate the graph files
from the SAM file, add additional files to data pool, edit the rings and annotation,
then render the image.

1. Open a new BRIG session.

2. Create the graph file from the graph files modules: Main window >Mod-
ules >Create graph files.

(a) Set drop down to coverage graph, fill in fields (Figure 7.1).
(b) Set Assembly file as “Mu50.sam” from the BRIG examples/Chapter7-
sam-examples folder.
(c) Set Output folder as the location of the Chapter7-sam-examples folder.
(d) Window size as “1”..
(e) Click Create Graph. This will add the graph file to the data pool when
it has finished.
7 VISUALISING GRAPHS AND GENOME ASSEMBLIES 27

Close the coverage graph window and return to the first main window.

1. Set reference file as “S.aureus.Mu50-plasmid-AP003367.gbk” from the Chapter7-

sam-examples folder. Users can use the browse button to traverse the file
system.

2. Set <unzipped BRIG examples folder>/Chapter7-sam-examples as the query

sequence folder.

3. Press “add to data pool”, this should load several items into the pool list.

4. Set the output folder as unzipped Chapter7-sam-examples folder.

5. Make sure the BLAST options box is blank.

Click next to move to the next window to configure the rings and add in annota-
tions.

1. Create 4 rings, name them “Mapping Coverage”, “pSK57”, “SAP014A”,

“CDS”.

2. Ring 1 Settings:

(a) Add “Mu50.sam.graph” from data pool to Ring 1.

(b) Set graph maximum value as “10”.
(c) Set colour as rgb(204,0,0).
(d) Set legend title as “Mapping coverage”.
(e) Check show red/blue.
7 VISUALISING GRAPHS AND GENOME ASSEMBLIES 28

3. Ring 2 Settings:

(a) Add “S.aureus.pSK57-plasmid-GQ900493.gbk” from data pool to Ring

2.
(b) Set colour as rgb(0,0,102).
(c) Set legend title as “pSK57”.

4. Ring 3 Settings:

(a) Add “S.aureus.SAP014A-plasmid-GQ900379” from data pool to Ring

3.
(b) Set colour as rgb(102,0,102).
(c) Set legend title as “SAP014A”.

5. Ring 4 Settings:

(a) Set colour as rgb(0,0,0).

(b) Set legend title as “CDS”.
7 VISUALISING GRAPHS AND GENOME ASSEMBLIES 29

Click “Add Custom features”.

1. Double-click Ring 4.
2. Set Input data to “Genbank”.
3. Set colour to “black”.
4. Set Draw feature as “default”.
5. Set Genbank file location to the location of
“S.aureus.Mu50-plasmid-AP003367.gbk”.
6. Set Feature as “CDS”
7. Click add.

This will load all the coding sequences from the Genbank file. These annotations
will be drawn as arrows, indicating orientation. Close this window and click next
on the second window.
1. Set title as “S. aureus Mu50 plasmid”.
2. Click Submit.
This will generate the final image, it should look like Figure 12.
7 VISUALISING GRAPHS AND GENOME ASSEMBLIES 30

Figure 12: S. aureus Mu 50 plasmid, showing read mapping from simulated 454
reads, CDSs, and genome comparisons to other S. aureus plasmids, pSK57 &
SAP014A. Alignments were performed with BLAST+
7 VISUALISING GRAPHS AND GENOME ASSEMBLIES 31

7.2 Walk through for visualising ace file assembly coverage.

This section will give a worked example of producing a BRIG image showing
assembly coverage read from an ace file. The final image will look like Fig-
ure 13. This walk through requires BRIG examples.zip from the BRIG website
(http://sourceforge.net/projects/brig/files/). Unzip this some-
where convenient.
The general procedure is to first generate the graph files from the ace file, con-
vert the coverage information to reference sequence if necessary, add additional
files to data pool, edit rings and annotation, then render the image.
Draft genome sequences are often modified to be consistant with other infor-
mation (e.g genome scaffolding, PCR sequencing of gaps) after being initially
assembled. This may change the order and size of the final genome sequence
compared to the original assembly.
To show the read coverage from the assembly on the final sequence correctly
the “Convert graph” module within BRIG can be used to map the coverage infor-
mation from the ace file onto the new sequence.
This module can also be used to map read coverage from an assembly onto a
closely-related reference genome.
7 VISUALISING GRAPHS AND GENOME ASSEMBLIES 32

First, produce the coverage graph file based off the assembly (ace file):

1. Open a new BRIG session.

2. Create the graph file from the graph files module: Main window >Modules
>Create graph files.

3. Set drop down to coverage graph, fill in fields (Figure 7.2).

(a) Set Assembly file as “454-S.aureus.Mu30.ace” from the BRIGEXAMPLE2-

ace folder.
(b) Set Output folder as the location of the BRIGEXAMPLE2-ace folder.
(c) Set window size as “1”.

4. Click “Create Graph”. This will add the graph file to the data pool when it
has finished.
7 VISUALISING GRAPHS AND GENOME ASSEMBLIES 33

Next, map the coverage generated in the previous graph file to the modified genome
sequence.

1. Remain in the “Create custom graph” window: Main window >Modules

>Create graph files.

2. Set drop down to convert graph, fill in fields as below.

(a) Set Original sequence as “454AllContigs-S.aureus.Mu50.fna”.

(b) Set New sequence as “S.aureus.Mu50-plasmid-AP003367.fna”.
(c) Set graph file as “454-S.aureus.Mu50.ace.graph”.
(d) Set Output folder as the location of the BRIGEXAMPLE2-ace folder.
(e) Window size as “1”.

3. Click “Create Graph”. This will add the graph file to the data pool when is
has finished.
7 VISUALISING GRAPHS AND GENOME ASSEMBLIES 34

Close the “Create custom graph” window and return to the main window.

1. Set reference file as “S.aureus.Mu50-plasmid-AP003367.gbk” from the

BRIGEXAMPLE2-ace folder. Users can use the browse button to traverse
the file system.

2. Set <unzipped BRIGEXAMPLE2-ace folder>/genomes as the query se-

quence folder.

3. Press “add to data pool”, this should load several items into the pool list.

4. Set the output folder as unzipped BRIGEXAMPLE2-ace folder.

5. Make sure the BLAST options box is blank.

7 VISUALISING GRAPHS AND GENOME ASSEMBLIES 35

Click next to move to the next window and configure the rings.

1. Create 4 rings.

2. Ring 1 Settings:

(a) Add “454AllContigs-S.aureus.Mu50.fnanew.graph” to Ring 1.

(b) Set colour as rgb(204,0,0).
(c) Set legend title as “Assembly coverage”.
(d) Set graph maximum value as “700”.
(e) Check show red/blue.

3. Ring 2 Settings:

(a) Add “S.aureus.pSK57-plasmid-GQ900493.gbk” to Ring 2.

(b) Set colour as rgb(0,0,102).
(c) Set legend title as “pSK57”.

4. Ring 3 Settings:

(a) Add “S.aureus.SAP014A-plasmid-GQ900379” to Ring 3.

(b) Set colour as rgb(102,0,102).
(c) Set legend title as “SAP014A”.

5. Ring 4 Settings:

(a) Set colour as rgb(0,0,0).

(b) Set legend title as “CDS”.
7 VISUALISING GRAPHS AND GENOME ASSEMBLIES 36

The rings are now step up with the correct colour, data and labels. The next step
is to mark the CDS on Ring 4 as “custom features”. Click “Add Custom features”
to open a new window.

1. Double-click on Ring 4.

2. Set Input data to “Genbank”.

3. Set colour to “black”.

4. Set Draw feature as “default”.

5. Set GenBank file location to “S.aureus.Mu50-plasmid-AP003367.gbk” in

the BRIGEXAMPLE2-ace folder.

6. Set Feature as “CDS”

7. Click add.
7 VISUALISING GRAPHS AND GENOME ASSEMBLIES 37

This will load all the coding sequences from the GenBank file. These annota-
tions will be drawn as clockwise or counter-clockwise arrows, indicating ori-
entation. Close this window and click next on the main window. From the
third/confirmation window:

1. Set title as “S. aureus Mu50 plasmid”.

2. Click submit.

This will generate the final image, which should look like Figure 13.
7 VISUALISING GRAPHS AND GENOME ASSEMBLIES 38

Figure 13: S. aureus Mu 50 plasmid published genome sequence, showing as-

sembly coverage from a draft assembly based on simulated 454 reads. The image
also shows CDS, and genome comparisons to other S. aureus plasmids, pSK57 &
SAP014A. Alignments were performed with BLAST+.
8 WALKTHROUGHS ON CREATING CUSTOM ANNOTATIONS 39

8 Walkthroughs on creating custom annotations

8.1 Adding custom annotations from a tab-delimited file, Gen-
Bank or EMBL file
BRIG can add features from a tab-delimited file, Multi-FASTA or GenBank/EMBL
file. This section deals with GenBank/EMBL & tab-delimited files. A walk-
through for generating annotations from a Multi-FASTA file can be found in Sec-
tion 6.2 on page 19.

8.1.1 Step 1: Load in sequences

The walk through will work out of the unzipped BRIG examples.zip folder. The
walk through and related figures will use C:\BRIG examples\Chapter5 6 8 whole
GenomeExamples as that location.
1. First, set “E coli O157H7Sakai.gbk” as the reference sequence from the
BRIGEXAMPLE folder.
2. Set <unzipped BRIG examples folder>\Chapter5 6 8 whole GenomeEx-
amples as the query sequence folder.
3. Press “add to data pool”, this should load several items into the pool list.
4. Set the Chapter5 6 8 whole GenomeExamples folder as the output folder.
5. The BLAST options box should be left blank.

6. Click next
8 WALKTHROUGHS ON CREATING CUSTOM ANNOTATIONS 40

8.1.2 Step 2: Configure rings

The next step is to configure what information is shown on each of the concentric
rings in BRIG. There should be three rings, for each ring:

1. Set the legend text for each ring

2. Select the required sequences from the data pool and click on “add data” to
add to the ring list.

3. Choose a colour

4. Set the upper (70) and lower (50) identity threshold.

5. Click on “add new ring” and repeat steps for each new ring required.

The values required for each ring are detailed in the table below.
Legend text Required sequences Colour
K12 MG1665 E coli K12MG1655.fna 0,153,255
CFT073 E coli CFT073.fna 204,0,255
UTI89 E coli UTI89.fna 153,0,102
8 WALKTHROUGHS ON CREATING CUSTOM ANNOTATIONS 41

8.1.3 Step 3: Adding annotations

Adding annotations from GenBank files
This step adds annotations read from a Genbank/EMBL file. These files can be
searched for a specific feature type containing a specific value. This walk through
searched the E. coli O157:H7 Sakai genome for all CDSs that are annotated as
adhesins. From the “Configure Rings” window, click “Add Custom features”.
These are the steps to load the GenBank features:

1. Create a new ring (Ring 4) and double-click Ring 4 in the ring list.

2. Set “input data” to Genbank.

3. Set “colour” to red.

4. Leave “Draw feature as” default.

5. Leave label text as a blank.

6. Set Genbank file location as Chapter5 6 8 whole GenomeExamples folder\

E coli O157H7Sakai.gbk

7. Set “Feature name” as CDS

8. Set “And” as AND

9. Set “Text contains” as adhesin.

10. Click add.

11. Remain in the “Add Custom features” for the next step.
8 WALKTHROUGHS ON CREATING CUSTOM ANNOTATIONS 42

PRO TIP 15: The same process applies for EMBL files.

By default, CDS on the sense strand will be drawn as clockwise arrows and
counter-clockwise arrows for anti-sense. All other features will be shown as block
arcs, by default. This can be overriden by changing “Draw feature as” from “de-
fault”.

Adding annotations from Tab-delimited files

This step adds annotations of the E. coli O157:H7 Sakai prophage site (SP
Sites) on to the image, which will be read from a tab-delimited file (SP-Sites.txt,
which can be found in the Chapter5 6 8 whole GenomeExamples folder ).
To add annotate the image with the SP-Sites:

1. Create a new ring (Ring 5) and double-click Ring 5 in the ring list.

2. Set “input data” as Tab-delimited.

3. Leave “label” blank.

4. Set “colour” to black.

8 WALKTHROUGHS ON CREATING CUSTOM ANNOTATIONS 43

5. Set tab-delimited file location as Chapter5 6 8 whole GenomeExamples\

Sp-Sites.txt

6. Press Add.

If there are a lot of entries, this make take a few seconds. All the values should
show up in the left hand pane. Individual changes can be made by double clicking
and editing an entry. Users can load any kind of results or annotations into BRIG
using this approach.

8.1.4 Step 4: Review and submit

The last window allows us to change the BLAST options, the location of the
image file and set the image title, which will appear in the centre of the ring. For
the walkthrough we need to configure the third window as:

1. Set the image title as “E. coli O157:H7 Sakai”.

2. Press submit.

3. The image will be created in the specified output directory and should look
something like Figure 14.
8 WALKTHROUGHS ON CREATING CUSTOM ANNOTATIONS 44

Figure 14: Output image from tab-delimited annotations. This figure shows E.
coli O157:H7 Sakai as the central reference sequence, with genome comparisons
to K12, CFT07 and UTI89 and the SP sites annotated in black.
8 WALKTHROUGHS ON CREATING CUSTOM ANNOTATIONS 45

8.2 How to create tab-delimited files for BRIG

BRIG can load custom annotations or results from a tab-delimited file. A tab
delimited file; a plain text file that represents a table, with a tab between each
column in the text. Rows are on each line and columns are separated by a tab
character (called the delimiter). BRIG only supports tab-delimited files, not other
delimiters e.g commas, space etc.
For BRIG to accept custom annotation files, it must fulfill the following re-
quirements:

• Lines that contain column headers, or comments must start with a “#”. See
the first line in Figure 8.2.

• The first two columns MUST be Start and Stop values and they must have
a value or BRIG will ignore them.

• Specifying Labels, colours or decoration is not nessercary but if they must

be in the right column. (Label must be 3rd, Colour must be 4th, Decoration
must be 5th).

• Acceptable Colour values are: default, red, aqua, black, blue, fuchsia, gray,
green, lime, maroon, navy, olive, orange, purple, silver, teal, white, yellow.

• Acceptable Decoration values are default, arc, hidden, counterclockwise-

arrow, clockwise-arrow.

• Files must be in plain-text tab-delimited.

The easiest way to make BRIG tab-delimited files is to set up a spreadsheet with
exactly the same headers as below (Order and text is important) and then fill in
the columns with data, leave the field blank to use default colours and decorations.
Leave the label field blank if no label is required.
9 CONFIGURATION OPTIONS 46

9 Configuration options
9.1 Saving and reopening your work

Figure 15: File menu in BRIG

Users can save their work in BRIG from the File menu (Figure 15) and can
open these save files from File menu >Open.... There are three options for sav-
ing:
• “Save As...” will save all the image settings, BRIG settings, rings configu-
ration, files users have imported into BRIG.

• “Save Profile...” will save just the global image settings & BRIG settings
like image dimensions and font size and colours. It will not save the data
pool or any ring settings. These files are designed to be used as templates
for other images.

• “Bundle Session...” will copy all files used in the BRIG session and save all
the current settings into a single directory. This directory can be compressed
and kept as an archive or sent to someone else who can open the .xml file
and work on the image with all of the original settings.

9.2 BLAST options

BRIG will run with default BLAST options if the BLAST options field is left
blank. BLAST options field (Figure 16) can be used to add custom BLAST op-
tions.

PRO TIP 16: BRIG handles all the file input and output into BLAST, so DO
NOT use -o, -d, -p, -i, -m in BLAST legacy or -out, -db, -query, -outfmt in
BLAST+.
9 CONFIGURATION OPTIONS 47

The list below highlights some BLAST parameters for BLAST+[2] and BLAST
legacy[1]; more common Parameters are in bold. For more information please
read the blastall or BLAST+ command-line usage messages (or http://www.
ncbi.nlm.nih.gov/books/NBK1763/).
BLAST+ options:

• -evalue Expectation value (E) [Real] default = 10.0

• -dust or -seg Filter query sequence (DUST with blastn, SEG with oth-
ers) [String] default = yes

• -task Task to execute. Permissible values: ’blastn’,’blastn-short’,’dc-

megablast’,’megablast’,’vecscreen’. Default for BLAST+ is ‘megablast’.
BRIG overrides this by forcing blastn, unless the user specifies otherwise.

• -num threads Number of processors to use [Integer] default = 1

• -html Produce HTML output [T/F] default = F

• -gilist Restrict search of database to list of GI’s [String] Optional

PRO TIP 17: BRIG runs BLAST+ with task as blastn, unless overidden by
the user.

BLAST legacy options:

• -e Expectation value (E) [Real] default = 10.0

• -F Filter query sequence (DUST with blastn, SEG with others) [String]
default = T

• -a Number of processors to use [Integer] default = 1

• -T Produce HTML output [T/F] default = F

• -l Restrict search of database to list of GI’s [String] Optional

• -W Word size, default if zero (blastn 11, megablast 28, all others 3) [Integer]
default = 0

• -n MegaBlast search [T/F] default = F

9 CONFIGURATION OPTIONS 48

Figure 16: Set custom BLAST options on the first window (shown) or the third
window

9.3 Setting BRIG options

The BRIG options window under Preferences>BRIG options in the first main
screen allows users to set options that include valid Genbank, EMBL or FASTA
file name extensions; percentage identity thresholds; and memory allocated to
CGView.
Changes can be applied to just the current session through “Save & Close”
or changes can be saved as the default settings for every new session in BRIG
through “Save settings as default”. There are two BRIG option tabs.

Tab 1 (Figure 17) has the following options:

• Genbank file extensions: Users can specify the file extensions they use as
GenBank files (with commas between extensions). e.g .gbk.

• FASTA file extensions: Users can specify the file extensions they use as
FASTA files (with commas between extensions). e.g .fa,.fna,.fas.

• EMBL file extensions: Users can specify the file extensions they use as
EMBL files (with commas between extensions). e.g .embl.
9 CONFIGURATION OPTIONS 49

• BLAST binary folder: Users must specify the location of BLAST executa-
bles, leave this blank if BLAST is on their PATH.

• Default spacer space: Users can set a default spacer value that BRIG uses
when using Multi-FASTA files.

Figure 17: First BRIG options window, accessible from Preferences >BRIG
options

Tab 2 (Figure 18) has the following options:

• Graph label thickness: Graph rings has another smaller ring to show la-
belled regions, users can set how many times smaller these rings are com-
pared to normal rings here.

• Graph ring thickness multiplier: Graph rings need to be larger than other
rings, users can set how many times larger Graph rings are compared to
normal rings here.

• Image render memory (MB): Amount of memory give to CGView for

image rendering, if users run out of memory when rendering thier image
they should increase this value.

• Default upper & lower threshold: The colour of BLAST matches in BRIG
are shaded in a sliding scale between 100% and the lower threshold percent-
age identity. These values and their corresponding colour are shown in the
legend. The upper threshold is used to highlight a particular percentage
9 CONFIGURATION OPTIONS 50

Figure 18: Second BRIG options window, accessible from Preferences >BRIG
options

identity in the legend and has no bearing on the colour of BLAST matches.

• Default min. threshold: Minimum percent identity for BLAST results that
BRIG will include on an image.

9.4 Setting Image options

Image rendering is done through CGView[5], these settings can be changed from
Preferences >Image options. A detailed explanation of these options can be
found at http://wishart.biology.ualberta.ca/cgview/xml\_cgview.
html.
Here is list of some of the more useful options:

• “Global settings” tab

– Height and Width:These values change the height and width of image
by specifying the number of pixels.
– Title font and color: These values change the font type, size and
colour.
– Show shading?: Set to false to turn off the embossing on annotations.
9 CONFIGURATION OPTIONS 51

• “Feature settings” tab

– FeatureSlot spacing: Sets the amount of gaps between rings.

– Feature thickness: This sets the default thickness of rings on image.
This can be overidden by the thickness parameter on ring customiza-
tion window.

• “Tick settings” tab

– Tick density: This value changes how often ticks appears on the ruler,
requires a decimal number between 0 and 1.
– Long tick colour: This value changes the colour of long (or short
ticks).

• “Backbone and ruler settings” tab

– Backbone radius: Sets the radius size of the ring circle. Increase this
to make a larger ring and decrease to make it smaller. N.B this will not
automatically scale with the Image height and width.

• “Label settings” tab

– Labels on?:Global settings for whether feature labels should be drawn

on this image.
– Global label colour: Default colour of labels, will be overriden with
individual settings.
– Label font: Font of all labels, set globally.
– Labels to keep: Number of labels to try and keep on the image.

• “Legend & Warning font settings” tab

– Legend and Warning font: Font of all legend and warning messages,
set globally.
9 CONFIGURATION OPTIONS 52

9.5 Loading a preset image template

BRIG has a set of sample templates the can be used for optimal dimensions and
font sizes for a given image. This can be accessed from Preferences >Load
template. BRIG will show a drop down of all the template in the profiles folder
in the BRIG installation folder, and a preview of the template (Figure 19), from
this users can choose the best template for their image and make adjustments in
“Image options”.
Users can add their own templates by saving a BRIG session as a profile and
copying it to the profile folder in their BRIG directory. To add a thumbnail for a
profile, copy a BRIG image for that session into the profile folder as a JPEG with
the same name as that profile. (i.e profilename.jpg).

PRO TIP 18: Loading a preset template will not overwrite the ring colour
settings. This is only to get the right proportions for image sizes, fonts and
ring sizes.

Figure 19: Loading in a preset template window

REFERENCES 53

Availability and requirements

Project name: BLAST Ring Image Generator (BRIG)
Project home page: http://sourceforge.net/projects/brig/
Operating system(s): Platform independent
Requirements: Java 1.6 or greater
Licence: GPLv3

References
[1] A LTCHUL , S., G ISH , W., M ILLER , W., M YERS , E., AND L IPMAN , D. Ba-
sic Local Alignment Search Tool. Journal of Molecular Biology 215, 3 (OCT
5 1990), 403–410.

[2] C AMACHO , C., C OULOURIS , G., AVAGYAN , V., M A , N., PAPADOPOU -

LOS , J., B EALER , K., AND M ADDEN , T. L. BLAST plus : architecture and
applications. BMC BIOINFORMATICS 10 (DEC 15 2009).

[3] DARLING , A., M AU , B., B LATTNER , F., AND P ERNA , N. Mauve: Multiple
alignment of conserved genomic sequence with rearrangements. GENOME
RESEARCH 14, 7 (JUL 2004), 1394–1403.

[4] R ICHTER , D. C., OTT, F., AUCH , A. F., S CHMID , R., AND H USON , D. H.
MetaSim-A Sequencing Simulator for Genomics and Metagenomics. PLOS
ONE 3, 10 (OCT 8 2008).

[5] S TOTHARD , P., AND W ISHART, D. Circular genome visualization and ex-
ploration using CGView. BIOINFORMATICS 21, 4 (FEB 15 2005), 537–539.