SWEEP

Sliding Window Extraction of Explicit Polymorphisms

This is a free pipeline released under the MIT license (See included file LICENSE for details). For allopolyplod species will filter allelic SNPs from homeologous SNPs (false positives). Validated for 85% accuracy rate by Sanger Sequencing.

written by Josh Clevenger 2014

REQUIRED:
-b sorted/indexed bam file of alignments
-g indexed genome
-o filtered vcf output name

OPTIONAL:
-s genotypic liklihood filtering stringency: default (0): 0 -> low, 1 -> medium, 2 -> high

-d minimum read depth filter per genotype: default (4): 1 - 100
-r minimum ratio of alternate allele to reference allele: default (0): 0 - 2
-w window size in bp: default (100) should be <= read length for optimal quality
--no_cleanup does not delete intermediate vcf files: default (FALSE)
--ultimate performs ultimate filtering for all homozygous calls checks all reads mapped to base for any alterate allele (will take longer)
REQUIRES Biopython with pysam module installed in your path!!

-vcf Optional use an existing vcf file as input
-num_genotypes if vcf file is used, enter number of genotypes

To use this script there are certain requirements:

(1) Samtools and Bcftools must be in your path -> Updated to be compatible with latest versions
(2) To do this easily you can add the path in your shell script
(3) alignment Bam files must be sorted and indexed
(4) Include as many bam files as you need separated with '-b'
(5) Option to include previously generated vcf file for SWEEP filtering (6) If input is vcf, you must still include bam files and genome for --ultimate filtering (7) Example command line:

  perl SWEEP.pl -b gen1.sorted.bam -b gen2.sorted.bam -g genome.fa -o output.vcf -s 1 -d 5 -r 0.25              

(8) Example with input vcf:

  perl SWEEP.pl -b gen1.sorted.bam -b gen2.sorted.bam -g genome.fa -o output.vcf -s 1 -d 5 -r 0.25 -vcf snps.vcf    
                -num_genotypes 2