#16 · Genomics

Population genomic inference of selection and ancestry from partial or linked data

The dominant thread is population-genetic inference from genomic data, though the four items target different quantities rather than a single model class. An ARG review assesses samplers for simulation and inference on performance, usability and biological realism. Two selection studies work from temporal data: an ancient DNA analysis of 7,244 people reports 34 selection signals using admixture-aware time-series and GWAS-based polygenic tests, while a linkage-aware pipeline infers fitness from short-read time-series by modelling linkage among partially observed allele pairs. FRAME estimates global ancestry composition from IBD-derived partial local ancestry with reference panel refinement, reported to achieve comparable accuracy. What is new is explicit treatment of incomplete or linked information, namely partial allele pairs and partial local ancestry, as a route to selection and ancestry inference.