search

arq5x / lumpy-sv

lumpy: a general probabilistic framework for structural variant discovery
MIT License
309 stars 118 forks source link

lumpyexpress on tumor/normal pair #267

Open arkyl opened 6 years ago

arkyl commented 6 years ago

Hi, I am wondering if it's possible to run tumor/normal pair with a simple lumpyexpress command line: lumpyexpress -B tumor.bam,normal.bam -o tumor-normal.vcf -x btu356_LCR-hs38.bed. So that the split-read and discordant-pair bams were not explicitly used in the command line. I was assuming that lumpyexpress itself would generate those proper files within its script pipeline? Thanks much.

Yue

ryanlayer commented 6 years ago

That should work.

On Aug 30, 2018, at 3:32 PM, ylark notifications@github.com wrote:

Hi, I am wondering if it's possible to run tumor/normal pair with a simple lumpyexpress command line: lumpyexpress -B tumor.bam,normal.bam -o tumor-normal.vcf -x btu356_LCR-hs38.bed. So that the split-read and discordant-pair bams were not explicitly used in the command line. I was assuming that lumpyexpress itself would generate those proper files within its script pipeline? Thanks much.

Yue

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.

arkyl commented 6 years ago

Hi, It works. Previously there were some problems but I think it's probably due to memory issues. Another question is about the output. I see there are SVs from both tumor and normal in the output vcf. What is the general down stream pipeline to generate the somatic SVs? I can imagine something like only retaining the "pass" SVs that are in tumor but not in normal. But considering tumor impurities, it seems to be more complicated than that. Any suggestions? Thanks a lot!

Yue

ryanlayer commented 6 years ago

Impurity is a complicated problem and there are quite a few methods out there that deal directly with this issue.

If you can assume that the normal has no tumor and the tumor has some normal, then try removing any SV that is in normal.

On Tue, Sep 4, 2018 at 8:36 AM ylark notifications@github.com wrote:

Hi, It works. Previously there were some problems but I think it's probably due to memory issues. Another question is about the output. I see there are SVs from both tumor and normal in the output vcf. What is the general down stream pipeline to generate the somatic SVs? I can imagine something like only retaining the "pass" SVs that are in tumor but not in normal. But considering tumor impurities, it seems to be more complicated than that. Any suggestions? Thanks a lot!

Yue

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/arq5x/lumpy-sv/issues/267#issuecomment-418390889, or mute the thread https://github.com/notifications/unsubscribe-auth/AAlDUcqvFacXCWa9S0iXLU89CQDERaN1ks5uXo_bgaJpZM4WUQWe .

arkyl commented 6 years ago

Thanks much for the suggestions! Yue