Integrative modeling of read depth and B-allele frequency improves single-cell copy number calling from targeted DNA sequencing panels

Cancer cells frequently gain or lose copies of DNA segments, and detecting these changes in individual cells is critical for understanding how tumors evolve and resist treatment. A technology called Tapestri sequences DNA from thousands of single cells and produces two types of signals: how much DNA is present (read depth) and which version of each gene a cell carries (allele information). Existing tools mainly use the first signal, potentially missing important changes that only the second signal can reveal. We developed scPloidyR, a statistical method that combines both signals to more accurately identify DNA copy number changes in single cells. Through simulations and analysis of real cancer cell data, we found that using both signals together substantially improves detection when allele information is available - even a small amount of allele data makes a meaningful difference. However, when allele information is absent, the simpler depth-only approach performs better. Our work provides researchers with a new tool and practical guidance on when each approach is most effective for studying cancer at single-cell resolution.