Hello everyone,

I'm encountering an issue where MuSE is producing an abnormally low number of somatic SNV calls when processing my batch of WES tumor-normal paired samples. I would greatly appreciate any insights.

Control Test: I first ran MuSE on a reference dataset (e.g., a TCGA sample). The number of SNVs called by MuSE was comparable to that from Mutect2, both falling within a reasonable range (e.g., 3,000-5,000). However, when I applied the exact same command and parameters to my own cohort of 50 breast cancer WES samples, MuSE called only between 14 to 100+ somatic SNVs per sample. This number is orders of magnitude lower than expected for WES data.

Running GATK Mutect2 (v4.3.0.0) on the same BAM files produces a normal, expected number of somatic calls (e.g., ~2,000-5,000 per sample). This confirms that the input data contains detectable somatic variants and the issue is specific to MuSE.

No Obvious Errors: The MuSE runs (call and sump steps) complete successfully without any error messages or warnings, making the root cause difficult to identify.

I would be very grateful for any advice or suggestions. I am happy to provide additional logs or file snippets if needed.