Detection of CNVs

Question

[nkausthu]

After calculating each exon coverage using Depth of Coverage module in GATK, how to normalize each exon depth and calculate copy number ratio?

Answer 1

[WouterDeCoster]

There are multiple possible solutions. I would suggest having a look at published algorithms, and a good place to start would be https://omictools.com/cnv-detection2-category

More precisely, XHMM starts from coverage depth calculated by GATK. But there are multiple other tools (exomeCopy, codex, conifer,...)

[nkausthu]

Thanks a lot and XHMM seems to a good option for me..

Answer 2

[chen]

The key point of CNV detection is you have to make a baseline from controls, and do the normalization based on the baseline.

[nkausthu]

As I want to find the CNVs from exomes of patients with Mendelian inheritance, any normal individual exome can be taken as a control? Does it make sense when two different patients are taken as control and normal?

[chen]

You can use several healthy people's exomes as control.

[WouterDeCoster]

But the data has to be obtained by the same library prep (kit) and sequencer to be comparable.

[chen]

Yes, so if you want to call CNV accurately, you have to sequence some healthy people with the same kit, using the same wet lab protocols.

[nkausthu]

Thank you so much ...