I've ran picard's CollectSequencingArtifactMetrics on my WES samples, both on my fresh-frozen tumors and blood normals.
In trinucleotide context, I'm seeing high "bait bias" errors CG>CA and CG>TG in both tumors and normals. These two error types cluster separately from other trinucleotide errors.
Is this something I should be alarmed about?
OxoG, FFPE metrics, Cref and Gref metrics are fine.
This is my first time working with WES data in general.