scEvoSel maps evolutionary selection signals onto single-cell transcriptomic atlases to reveal selection-relevant cellular contexts.

scEvoSel maps DCMS-based selection signals onto single-cell atlases, identifying fibro-adipogenic progenitors and myocytes in wild boar skeletal muscle and hepatocytes and Kupffer cells in Leghorn liver.

• Why it matters: Genome-wide scans find selection-affected genomic regions, but linking those signals to the specific cell types where they act remains difficult.
• What they did: The framework computes DCMS scores from population variation in wild boars vs domestic pigs and maps them to a wild boar skeletal muscle single-cell atlas and a Duroc pig atlas; it repeats this for red junglefowl vs Leghorn chickens mapped to a Leghorn liver atlas.
• The result: It accurately identifies selection-relevant cell types with significant cell-proportion changes (FAPs, myocytes) and breed-associated transcriptional divergence (lymphoid cells), and it highlights hepatocytes, endothelial cells, Kupffer cells, and B cells, enabling selection-linked genes and pathways in those contexts.