variant filtering of mouse whole exome sequencing

[rzeng]

hi, I have general question here about exome sequencing variant filtering.
I have generated mouse exome variants data from several samples using my pipeline including BWA alignment, samtools, picard and then GTAK. Also i got variant annotation table using Annovar.

However, I was struck here since I have biology background with not computer program skills before. First I need to filter variants by using public resource dbsnp or variant data. I have downloaded variant data from Sanger which has variants callings from 17 mouses. however, i do not know how to start to filter these data using it. I have tried with Annovar variant_deduction.pl but it did not work on non-human data.

Second, I have variant data from several different samples but how i use these data to filter variants? Is there good open resource that I just need to input my variant data and it can help me to filter these data?

http://doc.goldenhelix.com/SVS/latest/index.html

I found this SNP & Variation Suite from Golden Helix website. Anyone has used it and if it works just for human data too?

Thanks

[bw.]

You can use the GATK VariantAnnotator .

So something like:

java -jar GenomeAnalysisTK.jar -T VariantAnnotator
-R mm9.fa
--variant input.vcf
--dbsnp dbsnp135_path.vcf
-A DepthPerAlleleBySample # you can run with --list to see other -A options
-resource:my_dbsnp,vcf dbsnp135_path.vcf -E my_dbsnp.infofieldname
-resource:my_snps2,vcf my_snps2_path.vcf -E my_snps2.infofiledname1 -E my_snps2.infofieldname2 -comp my_snps2, vcf my_snps2_path.vcf
-o output.vcf


This will produce output.vcf which has all the variants in input.vcf along with additional INFO field values based on values from the other .vcf files.
It should then be possible to use GATK SelectVariants (or some other filtering tool) to extract the variants that you want.

GATK tools can be somewhat complicated to use, so there's also this list:
http://nickloman.github.io/2013/01/1...ing-vcf-files/

[rzeng]

BW. thank you for the information!

Is -A: allelic depths which is the number of AD like example below that allelic depth is 0+2 =2 then I can set -A as a number based on the AD?

GT: AD: DP:GQ:PL 1/1:0,2:2:6:66,6,0


Also I do not quite understand what was the -resource my_dbsnp,vcf dbsnp135_path.vcf -E my_dbsnp.infofieldname -resource:my_snps2,vcf my_snps2_path.vcf -E my_snps2.infofiledname1 -E my_snps2.infofieldname2 -comp my_snps2, vcf my_snps2_path.vcf exactly mean?

Suppose that my sample variant data is input1.vcf and I have other 5 extra variant data as input2/input3/input4/input5/input6.vcf and the i type
--variant input1.vcf and type -resource like follows??

-resource input2.vcf input2_path.vcf -E input2.infofildname
-resource input3.vcf input3_path.vcf -E input3.infofildname
.........

What is -comp my_snps2 mean? I could not find a clear explain for this in GATK annotation argument details.

Originally Posted by bw.

You can use the GATK VariantAnnotator .

So something like:

java -jar GenomeAnalysisTK.jar -T VariantAnnotator
-R mm9.fa
--variant input.vcf
--dbsnp dbsnp135_path.vcf
-A DepthPerAlleleBySample # you can run with --list to see other -A options
-resource:my_dbsnp,vcf dbsnp135_path.vcf -E my_dbsnp.infofieldname
-resource:my_snps2,vcf my_snps2_path.vcf -E my_snps2.infofiledname1 -E my_snps2.infofieldname2 -comp my_snps2, vcf my_snps2_path.vcf
-o output.vcf


This will produce output.vcf which has all the variants in input.vcf along with additional INFO field values based on values from the other .vcf files.
It should then be possible to use GATK SelectVariants (or some other filtering tool) to extract the variants that you want.

GATK tools can be somewhat complicated to use, so there's also this list:
http://nickloman.github.io/2013/01/1...ing-vcf-files/