Comparing exome sequencing data sample with vastly different read coverage

Question

[matthew.1.davies]

I have a question about exome sequencing data. I have sample of data from the same individual taken from blood and tumour samples. The read coverage of the tumour sample is ~500X but for the blood sample it's ~100X. This coverage is fine for calling somatic mutations but in comparing the sample for CNV calls, is it a concern that the difference between read coverages is so vast? While the coverage seems adequate for both samples individually, is it possible the difference in read count will distort the estimated log ratios between the samples when they differ by several magnitudes.

Thank you for any advice you can give.

Matthew

Answer 1

[Sean Davis]

The answer may depend on the algorithm you are using for CNV calling but, in general, most algorithms account for differing coverage between samples by "centering" the resulting copy number estimates in some way. In other words, the differences in coverage is likely not going to be problematic.

[matthew.1.davies]

Thank you, Sean.