speed up options

[ramaniak]

Hello, I am using svtyper on the results of a lumpy somatic vcf and am therefore running it with 2 bwa-mem bams and 2 split reads bam files. I find that the progess is very slow at this point : ~200 variants out of 9000 have been genotyped after 20hrs. Just wondering if there were any flags or ways to speed up the process other than splitting the vcf into smaller chunks.

thanks Arun

[CBorreda]

I've been splitting the VCF when I needed to speed things up. If you split it per sample you shouldn't have any problem, but if you try to split it per window (ie 3000 variants each) note that, for BND type SVs, both events have to be in the same file. Those are identified by their ID (x_1 and x_2), and they have to be in the same file, otherwise they will be deleted from your final output.

El 5 dic. 2017 10:54 a. m., "Arun Ramani" [email protected] escribió:

Hello, I am using svtyper on the results of a lumpy somatic vcf and am therefore running it with 2 bwa-mem bams and 2 split reads bam files. I find that the progess is very slow at this point : ~200 variants out of 9000 have been genotyped after 20hrs. Just wondering if there were any flags or ways to speed up the process other than splitting the vcf into smaller chunks.

thanks Arun

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/hall-lab/svtyper/issues/69, or mute the thread https://github.com/notifications/unsubscribe-auth/AZ8ErNFYc8Og1nn0kk8ZA9JEHSO7y6Bcks5s9ZFagaJpZM4Q2yWj .

[ramaniak]

Thank for the quick response. I will split by sample and try that out.

thanks Arun