I’m trying to better understand how to interpret borderline copy-number shifts in low-coverage whole-genome sequencing CNV analysis.

Context:
This is low-pass WGS used for copy number detection from a very small DNA sample (similar to single-cell / few-cell workflows). The pipeline reports copy number estimates and QC metrics such as WavSD and MAPD.

Observed pattern:

• Several chromosomes show copy number estimates around 2.25–2.4
• Multiple chromosomes affected (e.g., 1, 2, 9, 17, 22)
• QC metrics appear good (WavSD 0.10, MAPD 0.21)
• CNV calls flagged as low confidence by the pipeline

Questions:

1. In low-coverage CNV workflows, how often do systematic normalization or amplification biases produce similar intermediate copy numbers across multiple chromosomes?

2. Are there signal characteristics that help distinguish:

   • true mosaic copy-number variation
   • from GC-bias / amplification artifacts / segmentation drift?

3. When copy numbers cluster around ~2.3 across several chromosomes, is that typically interpreted as a shared abnormal lineage (biological mosaicism), or more commonly as a normalization artifact?

If helpful, I can share an anonymized screenshot of the CNV table.

I’m mainly trying to understand how bioinformaticians evaluate borderline CNV amplitude signals in these kinds of datasets.