simple_sv_annotation.py

A tool for simplifying snpEff annotations

Table of Contents

1. Requirements
2. Usage
3. Alternate Exon Numbers
4. Supported SV Types
5. Supported SV Callers
6. Example Output

Requirements

1. python 2.7
2. PyVcf python module
3. VCF file annotated with snpEff v4.3+.

simple_sv_annotation.py is designed around the new ANN annotation field rather than the previous EFF field.

Usage

usage: simple_sv_annotation [options] vcf

Required arguments

vcf FILE - vcf file annotated with snpEff v4.1g+

Optional arguments

--output/-o   FILE - Output file name. Use dash (-) for stdout. Default: <invcf>.simpleann.vcf.
--exonNums/-e FILE - List of custom exon numbers (see Alternate Exon Numbers)
--gene_list/-g FILE - List of genes to prioritise on
--known_fusion_pairs/-k FILE - Comma delimited file with a gene pair on each row representing known fusion pairs

Licence

This program is distributed under the MIT licence save for the exception below.

The file fusion_pairs.txt provided here is an extract of the file at https://github.com/ndaniel/fusioncatcher/blob/master/bin/generate_known.py and is redistributed here under the GNU GPLv3.

Alternate Exon Numbers

Occasionally the exon numbering scheme provided by snpEff is incorrect. snpEff numbers the exons in a transcript sequentially, but sometimes the accepted exon numbering is not sequential. For example, BRCA1 transcript 1, NM_007294, does not have an exon 4.

simple_sv_annotation.py accepts a BED file in which a user can provide custom numbering for a particular transcript. If a variant is annotated with a transcript listed in this file, the exon numbers provided by snpEff are replaced with the exon numbers in the file. If a transcript is not in the file, then the snpEff exon numbers are used. Follow the format below, separating each field with a tab

chr17    41196311    41197819    NM_007294|24
chr17    41199659    41199720    NM_007294|23
chr17    41201137    41201211    NM_007294|22
chr17    41203079    41203134    NM_007294|21
chr17    41209068    41209152    NM_007294|20
chr17    41215349    41215390    NM_007294|19
chr17    41215890    41215968    NM_007294|18
chr17    41219624    41219712    NM_007294|17
chr17    41222944    41223255    NM_007294|16
chr17    41226347    41226538    NM_007294|15
chr17    41228504    41228631    NM_007294|14
chr17    41234420    41234592    NM_007294|13
chr17    41242960    41243049    NM_007294|12
chr17    41243451    41246877    NM_007294|11
chr17    41247862    41247939    NM_007294|10
chr17    41249260    41249306    NM_007294|9
chr17    41251791    41251897    NM_007294|8
chr17    41256138    41256278    NM_007294|7
chr17    41256884    41256973    NM_007294|6
chr17    41258472    41258550    NM_007294|5
chr17    41267742    41267796    NM_007294|3
chr17    41276033    41276132    NM_007294|2
chr17    41277287    41277500    NM_007294|1

In the fourth column, provide the transcript name followed by a "|" and then the exon number. Note that the transcript version is not used. You may have additional fields in the bed file, simple_sv_annotation.py will only consider the first four.

Note: currently this list of alternate exons is stored in memory because it is expected to be relatively small. Very large lists of alternate exon numbering may affect performance.

Supported SV Types

simple_sv_annotation.py will attempt to simplify interesting and easy SV types to make the annotation result more interpretable. If you have an additional SV type that you want to be able to simplify, please email David Jenkins, AZ Email or BU Email.

1. Intergenic SVs
2. Intronic SVs
3. Whole Exon Loss SVs
4. Gene Fusions (can result from BND/DEL/INV/DUP)

Examples of the simplified SV annotations are below.

Supported SV Callers

simple_sv_annotation.py has been tested on annotated vcf output files from the following SV callers:

1. lumpy-sv
2. manta
3. delly

Additional SV callers will also work with simple_sv_annotation.py if VCF specifications are followed and each SV is described with standard SV INFO fields:

1. SVTYPE
2. MATEID (for SVTYPE=BND)
3. END (for whole exon deletions)

Example Output

Primary output for simple_sv_annotation.py:

1. Add SIMPLE_ANN field

In the default mode, simple_sv_annotation.py will not alter the ANN field provided by snpEff. Instead an additional field called SIMPLE_ANN will be added to the SV call. A SIMPLE_ANN will only be added to variants that can be simplified, other variants are not altered.

There are six fields in the SIMPLE_ANN tag separated by "|".

1. SV type (deletion, duplication, insertion, breakend)
2. Annotation (fusion, exon loss, intergenic, intronic)
3. Gene name
4. The seventh SnpEff field (often transcript name)
5. For exon loss variants, deleted exon numbers (Exon5del). For fusions, one of KNOWN_FUSION, ON_PRIORITY_LIST or NOT_PRIORITISED
6. Priority of the event (1 highest, 3 lowest)

example:

before:

chr17  41258467  del_5  ATATACCTTTTGGTTATATCATTCTTACATAAAGGACACTGTGAAGGCCCTTTCTTCTGGTTGAGAAGTTTCAGCATGCAAAATCTATA  A  .  .  END=41258555;SVTYPE=DEL;SVLEN=-88;UPSTREAM_PAIR_COUNT=0;DOWNSTREAM_PAIR_COUNT=0;PAIR_COUNT=0;ANN=A|exon_loss_variant&splice_acceptor_variant&splice_donor_variant&splice_region_variant&splice_region_variant&splice_region_variant&splice_region_variant&intron_variant&intron_variant|HIGH|BRCA1|BRCA1|transcript|NM_007294.3|Coding|4/23|c.135-5_212+5delTATAGATTTTGCATGCTGAAACTTCTCAACCAGAAGAAAGGGCCTTCACAGTGTCCTTTATGTAAGAATGATATAACCAAAAGGTATA||||||

after:

chr17  41258467  del_5  ATATACCTTTTGGTTATATCATTCTTACATAAAGGACACTGTGAAGGCCCTTTCTTCTGGTTGAGAAGTTTCAGCATGCAAAATCTATA  A  .  .  END=41258555;SVTYPE=DEL;SVLEN=-88;UPSTREAM_PAIR_COUNT=0;DOWNSTREAM_PAIR_COUNT=0;PAIR_COUNT=0;ANN=A|exon_loss_variant&splice_acceptor_variant&splice_donor_variant&splice_region_variant&splice_region_variant&splice_region_variant&splice_region_variant&intron_variant&intron_variant|HIGH|BRCA1|BRCA1|transcript|NM_007294.3|Coding|4/23|c.135-5_212+5delTATAGATTTTGCATGCTGAAACTTCTCAACCAGAAGAAAGGGCCTTCACAGTGTCCTTTATGTAAGAATGATATAACCAAAAGGTATA||||||;SIMPLE_ANN=DEL|EXON_DEL|BRCA1|NM_007294.3|Exon5del