Annotation

Annotating variants using multiple annotation databases, a tutorial

1. Getting annotation databases

This tutorial demonstrates how to use various databases to annotate variants in a variant tools project. These databases will be automatically downloaded and saved in directory ~/.variant_tools when they are used in a project. The amount of time required to download these databases depends on the speed of your internet connection, server load, and size of the databases. If you do no want to wait for the downloads and if you have enough disk space, you can download all variant tools resources into your local resource directory using the following commands:

% wget --mirror http://vtools.houstonbioinformatics.org
% rm -rf ~/.variant_tools   # remove existing directory if exists
% mv vtools.houstonbioinformatics.org ~/.variant_tools

The amount of data to download is 29G as of October 2012, and is expected to grow over time. The --mirror option allows command wget to get all files recursively, skipping files that exist locally.

2. Download a snapshot project with some variants

Let us create a project and download a snapshot project called vt_quickStartGuide.

% vtools init anno


INFO: variant tools 2.0.0 : Copyright (c) 2011 - 2012 Bo Peng
INFO: San Lucas FA, Wang G, Scheet P, Peng B (2012) Bioinformatics 28(3):421-422
INFO: Please visit http://varianttools.sourceforge.net for more information.
INFO: Creating a new project anno




% vtools admin --load_snapshot vt_quickStartGuide


Downloading snapshot vt_quickStartGuide.tar.gz from online
INFO: Snapshot vt_quickStartGuide has been loaded

This project has variants from CEU and JPT populations of the 1000 genomes pilot study. As we can see from the following commands, it has 4,858 variants from two samples:

% vtools show samples


sample_name	filename
CEU	CEU.exon...3.sites.vcf.gz
JPT	JPT.exon...3.sites.vcf.gz




% vtools show tables


table                 #variants     date  message
variant                   4,858   

3. Annotating variants

3.1 Listing available annotation databases

These are available annotation databases (as of October 2012) that can be downloaded and used automatically to annotate variants within a variant tools project. You can use the following command to list currently available databases.

% vtools show annotations


CancerGeneCensus-20111215
CancerGeneCensus
ccdsGene-hg19_20110909
ccdsGene-hg19_20111206
ccdsGene
ccdsGene_exon-hg19_20110909
ccdsGene_exon-hg19_20111206
ccdsGene_exon
dbNSFP-hg18_hg19_1.1_2
dbNSFP
dbNSFP_light-hg18_hg19_1.0_0
dbNSFP_light
dbSNP-hg18_129
dbSNP-hg18_130
dbSNP-hg19_131
dbSNP-hg19_132
dbSNP
evs-hg19_20111107
evs
evs_5400
keggPathway-20110823
keggPathway
knownGene-hg18_20110909
knownGene-hg19_20110909
knownGene
knownGene_exon-hg18_20110909
knownGene_exon-hg19_20110909
knownGene_exon
phastCons-hg19_20110909
phastCons
refGene-hg18_20110909
refGene-hg19_20110909
refGene
refGene_exon-hg18_20110909
refGene_exon-hg19_20110909
refGene_exon
thousandGenomes-hg19_20110909
thousandGenomes

3.2 How do I add an annotation database to my project?

To add a gene-based annotation source such as ccdsGene to your project, the following command will accomplish this. If you haven't already downloaded this annotation database with this or another project, vtools will automatically download the database and associate ccdsGene annotations to your project.

% vtools use ccdsGene

3.3 What genes do my variants belong to?

There are several annotation sources that could be used to annotate your variants to gene transcripts. Some examples include refGene, knownGene and ccdsGene. To get more details of these databases use vtools show annotation ccdsGene -v2 (or a similar command with refGene or knownGene) as described previously. This command downloads the ccdsGene data source allowing variants to be annotated to transcripts.

3.4 What about the exon?

Gene-based annotation sources such as ccdsGene, refGene and knownGene have corresponding annotation sources that are exon-based: ccdsGene_exon, refGene_exon and knownGene_exon respectively (provided indirectly through the UCSC Genome Browser database). These exon-based annotation sources contain exon start and end coordinates that are used in lieu of gene start and end coordinates for linking the annotations to your variants.

% vtools use ccdsGene_exon

3.5 What pathways do my variants belong to?

This command downloads the keggPathway annotation source allowing variants to be annotated to KEGG pathways indirectly through transcript annotations (provided by the ccdsGene annotation source).

% vtools use keggPathway --linked_by ccdsGene.name


INFO: Downloading annotation database from http://vtools.houstonbioinformatics.org/annoDB/keggPathway.ann
INFO: Downloading annotation database from http://vtools.houstonbioinformatics.org/annoDB/keggPathway-20110823.DB.gz
: Unsupported scheme.
--18:54:24--  http://vtools.houstonbioinformatics.org/annoDB/keggPathway-20110823.DB.gz
           => `./keggPathway-20110823.DB.gz'
Resolving vtools.houstonbioinformatics.org... 70.39.145.13
Connecting to vtools.houstonbioinformatics.org[70.39.145.13]:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 350,847 [  /applications/x-gzip]

100%[=============================================================================>] 350,847      330.94K/s             

18:54:26 (329.78 KB/s) - `./keggPathway-20110823.DB.gz' saved [350847/350847]


FINISHED --18:54:26--
Downloaded: 350,847 bytes in 1 files
INFO: Using annotation DB keggPathway in project quickstart.
INFO: kegg pathway for CCDS genes

Now lets filter all of our variants to include only those involved in metabolic pathways. This command uses the pathway annotation source that we just downloaded to find all variants that are on transcripts of proteins known to be involved in metabolic pathways. These variants are then stored in a table called metabolic.

% vtools select variant 'kgDesc="Metabolic pathways"' -t metabolic                                                                                        
Running: 2,788 2.1K/s in 00:00:01
INFO: 109 variants selected.