How to normalised DNA sequencing depth for genomic variant discovery ?

Question

[AIMAR]

Hi everyone

I've some BAM files of DNA reads and I would like to normalize the reads depth before moving to variant calling for somatic data. Is there anyone who can help on this ? How to proceed or even any idea.

[zizigolu]

You don’t need to manually normalize read depth before variant calling. Most somatic variant callers (like Mutect2, Strelka2, VarScan2) already account for coverage variation, sequencing depth, and base quality in their statistical models.

[AIMAR]

Aaah I didn't know about that, could you share any document that mention that cause it'll be helpfull. Also the thing I've normal samples and tumors samples and there I wanna compare them . This is why I've tried to normalize the sequencing depth to be sure that all samples are kinda correct (no biaises, or outliers values ).