CNV analysis from exome sequencing data

Question

[mkamranazim]

Is it possible to confidently carry out CNV analysis using whole exome sequencing data? There are many papers now that suggest it is possible, but how confident can we be about the results? And when using older BAM files to carry out this CNV analysis, is there a possibility that newer tools will experience more errors because of incompatibility of data files and software?

[Titus]

I think it will depends on the quality of the library prepared. The coverage normalization part is crucial in CNV calling so to me replicates are as important as the generation of your exome sequencing data.