Dear Genome STRiP users,
I got the CN and CNQ value by SVCNVDiscovery. After sorting the CNQ and the corresponding CN value, I found some weird CN values and CNQ values as below -- 10 samples out of 10686 samples (24 samples: CNQ (not .) < 13, and 24 samples: 13 < CNQ < 20).
| ID | sample 1 | sample 2 | sample 3 | sample 4 | sample 5 |
|---|---|---|---|---|---|
| CNQ | . | . | . | 0.5 | 1.8 |
| CN | -1 | -1 | -1 | 2 | 2 |
So may I have your suggestions about:
What does CN = -1 and CNQ = . mean? Does it mean that there are no confident copy number values?
Whether 10 out of 10686 samples has -1 as CN value (24 samples: CNQ (not .) < 13, and 24 samples: 13 < CNQ < 20) is a good CNV calling or a bad calling?
If this is a bad calling, may I have your suggestion about how to improve it, such as running in multiple batches (but more samples will make the estimation better?), using longer -intervalList in SVCNVDiscovery, or using longer -L in the SVPreprocess (if really cannot load the whole 94million cram file)?
Thank you very much.
P.S. It looks like that the ordered bulletin in this markdown window does not work so my post does not have the number order at the head of each question.
Best regards,
Wusheng