Sparse Haplotype-Based Fine-Scale Local Ancestry Inference at Scale Reveals Recent Selection on Immune Responses

The study introduces SparsePainter and PBWTpaint, two efficient algorithms for fine-scale local ancestry inference, which significantly enhance computational performance for large genomic datasets. SparsePainter, in particular, is faster and more memory-efficient than traditional methods while maintaining high accuracy. Using data from 487,409 UK Biobank individuals and 2,504 from the 1000 Genomes Project, the study identified recent selection on immune response genes across seven ethnicities, highlighting genes like MRC1 and STAM. The research underscores the importance of local ancestry in understanding genetic history and selection, revealing that haplotype components (HCs) provide more detailed ancestry information than principal components (PCs). The findings suggest new insights into the historical interplay between pathogens and the immune system, with significant selection signals in immune-related genes.