verifybamID on tumors?

82 views
Skip to first unread message

Jo Lynne Harenza

unread,
Jun 5, 2018, 10:07:48 AM6/5/18
to verifyBamID
Hello,

We ran this method on some PDX tumors, which have already been mouse subtracted. However, about 40 of our 250 samples have FREEMIX values > 0.02. I am not sure if this is due to sample contamination or the nature of the sample. We do not have normals to compare to, but have seen some contaminated STRs (could be lab not sample issue) and I wanted some way to verify that our sequencing data was not contaminated. So, has anyone run this on tumor samples and if so, what is an acceptable value for FREEMIX? 

The fact that the majority came back < 0.02 makes me think that it should work on tumors and perhaps this is more of a residual mouse contamination? All other values are within expected ranges.

Thanks!

Jo Lynne

kiran girdhar

unread,
Jun 5, 2018, 10:18:10 AM6/5/18
to verif...@googlegroups.com
Hi Jo

what data type you have? I have similar issue with chip-seq data.

KG

--
You received this message because you are subscribed to the Google Groups "verifyBamID" group.
To unsubscribe from this group and stop receiving emails from it, send an email to verifybamid+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Jo Lynne Harenza

unread,
Jun 5, 2018, 12:05:37 PM6/5/18
to verif...@googlegroups.com
Hi Kiran,

This is WES data, but we have WES and RNA-Seq bams.

Jo Lynne
--

Hyun Min Kang

unread,
Jun 5, 2018, 3:00:35 PM6/5/18
to verif...@googlegroups.com
How many cases do you have FREEMIX greater than 5% or 10%?

Do you have array data for any of those samples to estimate CHIPMIX?

Do you expect the tumor samples have high number of genomic duplicates (which increase heterozygosity)? Usually LOH is much more frequent, so it should not inflate FREEMIX in my opinion. 

I have observed that some exome data had decent amount of FREEMIX values, but I couldn't recall what the exact reason was. Could you try the new verifyBamID at https://github.com/Griffan/VerifyBamID and see if it gives similar results?
…

On Wed, Jun 6, 2018 at 1:05 AM Jo Lynne Harenza <jhar...@gmail.com> wrote:
Hi Kiran,

This is WES data, but we have WES and RNA-Seq bams.

Jo Lynne
On Tue, Jun 5, 2018 at 10:17 AM, kiran girdhar <kiran.gi...@gmail.com> wrote:
Hi Jo

what data type you have? I have similar issue with chip-seq data.

KG

On Tue, Jun 5, 2018 at 10:07 AM, Jo Lynne Harenza <jhar...@gmail.com> wrote:
Hello,

We ran this method on some PDX tumors, which have already been mouse subtracted. However, about 40 of our 250 samples have FREEMIX values > 0.02. I am not sure if this is due to sample contamination or the nature of the sample. We do not have normals to compare to, but have seen some contaminated STRs (could be lab not sample issue) and I wanted some way to verify that our sequencing data was not contaminated. So, has anyone run this on tumor samples and if so, what is an acceptable value for FREEMIX? 

The fact that the majority came back < 0.02 makes me think that it should work on tumors and perhaps this is more of a residual mouse contamination? All other values are within expected ranges.

Thanks!

Jo Lynne

--
You received this message because you are subscribed to the Google Groups "verifyBamID" group.
To unsubscribe from this group and stop receiving emails from it, send an email to verifybamid...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "verifyBamID" group.
To unsubscribe from this group and stop receiving emails from it, send an email to verifybamid...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.
--

--
You received this message because you are subscribed to the Google Groups "verifyBamID" group.
To unsubscribe from this group and stop receiving emails from it, send an email to verifybamid...@googlegroups.com.
Reply all
Reply to author
Forward
0 new messages