Coverage anomaly?

[Kashliks]

Hello,

I am doing exome sequencing of non-malignant tumors on Ion Proton PI chip.
Sequencing works fine, but when

analysed BAM with Qualimap 2.0.2, I received strange picture of coverage:



and



while coverage of exome sequencing of leukocyte derived DNA is more even (except X):



Interesting that when analysing more closely there is high similarity between the tumor sample coverage - hills and valleys follow the same pattern across all samples (currently 3).

My aim is to find exome mutations that differ between tumor samples and blood (obviously). When analysed further tens of thousands SNVs are found different between blood derived DNA and tumor although literature suggests about 100-to 200 for this type of tumor.

Has anyone seen such pictures before? Are they normal? Can it be corrected and how?

Thanks in advance!