tumor_normal.vcf filtering

arq5x / lumpy-sv

lumpy: a general probabilistic framework for structural variant discovery

MIT License

316 stars 119 forks source link

tumor_normal.vcf filtering #249

Open chunlinxiao opened 6 years ago

chunlinxiao commented 6 years ago

Following the example (below) to generate tumor_normal.vcf file, it seems lumpy generated many more somatic SVs than other tools. Do you have any suggestion for filtering some low quality calls from the tumor-normal vcf file?

Run LUMPY Express on a tumor-normal pair lumpyexpress \ -B tumor.bam,normal.bam \ -S tumor.splitters.bam,normal.splitters.bam \ -D tumor.discordants.bam,normal.discordants.bam \ -o tumor_normal.vcf

thanks

ryanlayer commented 6 years ago

Did these get genotyped by SVTYPER?

On Wed, Jun 13, 2018 at 11:19 AM chunlinxiao notifications@github.com wrote:

Following the example (below) to generate tumor_normal.vcf file, it seems lumpy generated many more somatic SVs than other tools. Do you have any suggestion for filtering some low quality calls from the tumor-normal vcf file?

Run LUMPY Express on a tumor-normal pair lumpyexpress -B tumor.bam,normal.bam -S tumor.splitters.bam,normal.splitters.bam -D tumor.discordants.bam,normal.discordants.bam -o tumor_normal.vcf

thanks

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/arq5x/lumpy-sv/issues/249, or mute the thread https://github.com/notifications/unsubscribe-auth/AAlDUSc4T6Z2XKbJVAY0yqPw57TUKnbZks5t8UmTgaJpZM4Ummqh .

-- Ryan Layer

chunlinxiao commented 6 years ago

Not yet.

Just wanted to be sure, for running svtyper on tumor_normal.vcf, the -B option is for tumor_sample.bam, -S option for tumor_sample.splitters.bam, and the -i option for tumor_normal.vcf.

svtyper \
-B sample.bam \ -S sample.splitters.bam \ -i sample.vcf

sample.gt.vcf

thanks

chunlinxiao commented 6 years ago

or the -B and -S like:

-B tumor.bam,normal.bam -S tumor.splitters.bam,normal.splitters.bam

ryanlayer commented 6 years ago

Using the newest version from https://github.com/hall-lab/svtyper

Don't use -S and both bams to -B

On Wed, Jun 13, 2018 at 1:13 PM chunlinxiao notifications@github.com wrote:

or the -B and -S like:

-B tumor.bam,normal.bam -S tumor.splitters.bam,normal.splitters.bam

?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/arq5x/lumpy-sv/issues/249#issuecomment-397053119, or mute the thread https://github.com/notifications/unsubscribe-auth/AAlDUUbtXxRea9TDzk0gdjsWFEfN3tVDks5t8WRogaJpZM4Ummqh .

-- Ryan Layer

chunlinxiao commented 6 years ago

thanks Ryan.

after svtyper, any suggestion for the filtering? thanks.

chunlin

ryanlayer commented 6 years ago

Only consider variants that have sufficient depth (DP) in tumor and normal. This number will depend on your sequencing depth.
Remove all variants that have any alt (AO) evidence in the normal.
Keeps variants with enough depth in the tumor. This number will depend on your sequencing depth and sample purity.

On Tue, Jun 19, 2018 at 9:27 AM chunlinxiao notifications@github.com wrote:

thanks Ryan.

after svtyper, any suggestion for the filtering? thanks.

chunlin

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/arq5x/lumpy-sv/issues/249#issuecomment-398440633, or mute the thread https://github.com/notifications/unsubscribe-auth/AAlDUS-Y1eFjJk1onTakE3JMNsOmMgPUks5t-RhJgaJpZM4Ummqh .