Two questions: 1) filter first or merge first, 2) B allele freq figure for duplication CNVs

[xiaoqing-liu]

Dear Dr. Wang:
I have two questions after applying PennCNV to our twin data:

At the ‘annotation.md’ page (https://github.com/WGLab/PennCNV/blob/master/docs/user-guide/annotation.md), it listed the following two steps:
Filtering CNV calls by user-specified criteria
Merging adjacent CNV calls

My question is if we could conduct the merging first before filtering.

I ask this because we found that for one of our monozygotic twin pairs, one twin had a CNV at a locus but the other twin didn’t have this CNV. However, after we conducted merging first before filtering, the other twin also had a CNV at this locus.

My second question is about the figures. For one of the twins, PennCNV detected one duplication CNV at a locus. Is it correct that for duplications, the B allele frequency figure should have values at 0, 0.33, 0.67, and 1? We didn’t see this for this duplication CNV, instead we have the values 0 and 1 mostly but also some 0.5 at this region. Could it still be a valid duplication CNV without the 0.33 and 0.67 values for B allele freq?

Thank you in advance for your help with my questions!
Xiao-Qing

[kaichop]

Yes, it is best to merge before filtering. So two smaller CNVs are merged into the real longer one, and then it will not be removed due to being toosmall. If you show the figure, I can make a guess. It all depends on the actual length of region and the heterozygosity rate of that broader region (it sounds like a LOH region)

…

On Wed, Apr 5, 2023 at 11:49 AM xiaoqing-liu ***@***.***> wrote: Dear Dr. Wang: I have two questions after applying PennCNV to our twin data: At the ‘annotation.md’ page ( https://github.com/WGLab/PennCNV/blob/master/docs/user-guide/annotation.md), it listed the following two steps: Filtering CNV calls by user-specified criteria Merging adjacent CNV calls My question is if we could conduct the merging first before filtering. I ask this because we found that for one of our monozygotic twin pairs, one twin had a CNV at a locus but the other twin didn’t have this CNV. However, after we conducted merging first before filtering, the other twin also had a CNV at this locus. My second question is about the figures. For one of the twins, PennCNV detected one duplication CNV at a locus. Is it correct that for duplications, the B allele frequency figure should have values at 0, 0.33, 0.67, and 1? We didn’t see this for this duplication CNV, instead we have the values 0 and 1 mostly but also some 0.5 at this region. Could it still be a valid duplication CNV without the 0.33 and 0.67 values for B allele freq? Thank you in advance for your help with my questions! Xiao-Qing — Reply to this email directly, view it on GitHub <#106>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABNG3OAMKZ3HOHB7ZGHIYA3W7WH7RANCNFSM6AAAAAAWUIGQ4U> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[xiaoqing-liu]

Thank you so much for your reply, Dr. Wang! Here I attached the figures. 1) a CNV (in red and is shorter) from filtering first, 2) a CNV (much longer) from merging first. Thanks again! Xiao-Qing Xiao-Qing (Michelle) Liu 刘晓青 Principal Investigator and Senior Epidemiologist Department of Obstetrics, Gynecology, and Reproductive Sciences Department of Biochemistry and Medical Genetics Faculty of Health Sciences, University of Manitoba Co-Director, The Statistical Genetics and Bioinformatics Core Platform, Rady Faculty of Health Sciences Scientist, Children's Hospital Research Institute of Manitoba John Buhler Research Centre 807J - 715 McDermot Avenue Winnipeg, Manitoba Canada R3E 3P4 (204) 887-6986 ***@***.*** [cid:3251587d-626b-42b5-8df4-45a18e108c80] [cid:f352b6ee-f2d1-445f-a2d6-261279966719]

________________________________ From: Kai Wang ***@***.***> Sent: April 9, 2023 11:24 PM To: WGLab/PennCNV ***@***.***> Cc: Xiaoqing Liu ***@***.***>; Author ***@***.***> Subject: Re: [WGLab/PennCNV] Two questions: 1) filter first or merge first, 2) B allele freq figure for duplication CNVs (Issue #106) Caution: This message was sent from outside the University of Manitoba. Yes, it is best to merge before filtering. So two smaller CNVs are merged into the real longer one, and then it will not be removed due to being toosmall. If you show the figure, I can make a guess. It all depends on the actual length of region and the heterozygosity rate of that broader region (it sounds like a LOH region)

On Wed, Apr 5, 2023 at 11:49 AM xiaoqing-liu ***@***.***> wrote: Dear Dr. Wang: I have two questions after applying PennCNV to our twin data: At the ‘annotation.md’ page ( https://github.com/WGLab/PennCNV/blob/master/docs/user-guide/annotation.md), it listed the following two steps: Filtering CNV calls by user-specified criteria Merging adjacent CNV calls My question is if we could conduct the merging first before filtering. I ask this because we found that for one of our monozygotic twin pairs, one twin had a CNV at a locus but the other twin didn’t have this CNV. However, after we conducted merging first before filtering, the other twin also had a CNV at this locus. My second question is about the figures. For one of the twins, PennCNV detected one duplication CNV at a locus. Is it correct that for duplications, the B allele frequency figure should have values at 0, 0.33, 0.67, and 1? We didn’t see this for this duplication CNV, instead we have the values 0 and 1 mostly but also some 0.5 at this region. Could it still be a valid duplication CNV without the 0.33 and 0.67 values for B allele freq? Thank you in advance for your help with my questions! Xiao-Qing — Reply to this email directly, view it on GitHub <#106>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABNG3OAMKZ3HOHB7ZGHIYA3W7WH7RANCNFSM6AAAAAAWUIGQ4U> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

— Reply to this email directly, view it on GitHub<#106 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AIQ2AD2RGM4RW6NGYBYWZUTXAODRNANCNFSM6AAAAAAWUIGQ4U>. You are receiving this because you authored the thread.Message ID: ***@***.***>