Chatomics Field GuideWhat They Don't Teach You

Topic

Outlier samples in PCA

One sample sits far from the others in PCA; do I drop it?

An outlier in PCA is a question, not a verdict: it may be a failed library, a swap, a different tissue region, or the most interesting sample you have. The decision requires QC metrics and metadata, and dropping samples to make a result significant is p-hacking with extra steps.