What is the difference between CNV binning and aligning to gene?

Question

[QX]

Hi All,

I have datasets of Whole Genome Seq of cancer and non-cancer cell. I would like to investigate the copy-number variation between the two conditions. After aligning read to reference genome, there would be two possible ways to follow analysis:

1. Define the binning and count for read-depth per bin, which basically the copy-number variation analysis. Then, assign to genes that fall into those bins.
2. Align the reads into the genes themselves and count read depth for each gene.

I understand the first approach is common for CNV analysis. But does anyone know the difference between the two kind of analysis. Do we lose information on CNV if we use the second approach? For further integrate with RNA datasets, which approach is more sufficient to perform?