SimGEA: simulation-based method improves detection of locally adapted loci in population genomics

Researchers have posted a preprint on bioRxiv describing SimGEA, a new computational method for genotype–environment association (GEA) analysis. GEA approaches are used widely in evolutionary and population genetics to identify genomic loci where allele frequencies correlate with environmental variables — a signal taken as evidence of local adaptation.

A persistent methodological challenge in GEA is distinguishing genuine adaptive signals from spurious correlations that arise from population structure: if populations differ genetically for historical demographic reasons and also happen to inhabit different environments, many loci will falsely appear adaptive. Existing correction methods either reduce statistical power or produce elevated false-positive rates under certain conditions, a trade-off that limits their utility in species with complex demographic histories.

SimGEA addresses this by generating null distributions through explicit demographic simulations calibrated to the study population, then testing observed allele-frequency–environment correlations against those simulated expectations. The authors report that the approach outperforms existing methods on benchmarks involving structured populations.

The method is likely to be of primary interest to population geneticists, evolutionary biologists, and computational genomics researchers working on local adaptation, landscape genomics, or conservation genetics. The preprint has not yet been peer-reviewed.