Three questions about calling CNV in Germline.

Answered

Follow

rq m

• March 12, 2021 03:23
• Edited

Question 1:
I noticed that there is a sentence in the description of GermlineCNVCaller:

providing a cohort of germline samples sequenced using the same sequencing platform and library preparation protocol (in case of WES, the same capture kit).

I downloaded SRA data from two different articles, that is, I have two batches of data. What should I do?
Should I separate the two batches of data instead of mix of them together in process of DetermineGermlineContigPloidy and GermlineCNVCaller?
That is, should I do twice operations in the DetermineGermlineContigPloidy and GermlineCNVCaller process respectively?

Question 2:
What if the ploidy result of DetermineGermlineContigPloidy is known to be wrong by other means?
For example, the ploidy of a chromosome of certain sample is known to be 2 by means of flow cytometry or other ways, but the result given by DetermineGermlineContigPloidy is 3. Can I change the contig_ploidy.tsv result file from 3 to 2?

Question 3:
Does PostprocessGermlineCNVCalls generate VCF file only containing one sample information at a time?
How can I get a VCF file containing multiple sample information?

 

Thanks for reply.

0

• 

  Genevieve Brandt (she/her)

  • March 16, 2021 17:31

  Hi rq m,

  Here are the answers to your three questions, hope this helps!

  1. Yes, you should run these batches separately.
  2. Yes, you can change the ploidy in that file.
  3. Run GATK 4.2.

  Best,

  Genevieve

  0

  Comment actions Permalink
• 

  rq m

  • March 17, 2021 14:23
  • Edited

  Hi Genevieve Brandt.

  Thanks for your answer.
  I have understood question 1 and 2, but I still have doubts about question 3.
  I have used version 4.2. I called CNV according the tutorial (How to) Call common and rare germline copy number variants.

  In the end, I generated many VCF files. But, each VCF file contains only one sample CNV information. How can I get all samples CNV information in ONE VCF file, like calling SNP?

  0

  Comment actions Permalink
• 

  SkyWarrior

  • March 18, 2021 16:01

  There is a new tool called JointGermlineCNVSegmentation. There is no clear documentation available yet but you can try. I will be trying it in a few days. 

  0

  Comment actions Permalink
• 

  Genevieve Brandt (she/her)

  • March 19, 2021 04:31

  Yes, the JointGermlineCNVSegmentation tool was released in 4.2.0.0. There are release notes available here: https://github.com/broadinstitute/gatk/releases/tag/4.2.0.0

  Here are some notes as it was being built as well: https://github.com/broadinstitute/gatk/pull/6554

  1

  Comment actions Permalink
• 

  Archana Verma

  • August 27, 2021 09:11

  Hi, I have one confusion, how to change the ploidy of a chromosome to 2 from 3. 

  Is the example below is right?

  CONTIG_NAME	PLOIDY_PRIOR_0	PLOIDY_PRIOR_1	PLOIDY_PRIOR_2	PLOIDY_PRIOR_3
  I	0.01	0.98	0.01	0
  II	0.01	0.98	0.01	0
  III	0.01	0.98	0.01	0

  0

  Comment actions Permalink
• 

  Pamela Bretscher

  • August 30, 2021 14:05

  Hi Archana Verma,

  Yes, this should work. Setting the prior probability to 0 will remove that ploidy state and forbid the tool from assuming a ploidy of 3. There is additional information in the DetermineFermlineContigPloidy Tool Documentation. Please let me know if this answers your question.

  Kind regards,

  Pamela

  0

  Comment actions Permalink
• 

  Archana Verma

  • September 02, 2021 05:41

  Hi Pamela Bretscher,

  Thanks for your answer.
  I have understood the explanation.

  0

  Comment actions Permalink

Please sign in to leave a comment.