Preprint: STRUCTURE reliably detects contemporary hybridisation when tested against simulated conservation scenarios
A simulation study finds that the widely used STRUCTURE program correctly identifies at least one hybrid individual in 100% of tested scenarios when population differentiation is sufficient, offering reassurance for its use in conservation genetics.
A preprint posted to bioRxiv evaluates how well the STRUCTURE program — one of the most widely used tools in population genetics — performs at a specific, practically important task: detecting contemporary hybridisation between previously isolated populations, as might occur following human-caused range overlap or translocation in conservation contexts.
The authors simulated hybridisation between two populations across timescales comparable to documented real-world cases, then tested STRUCTURE's ability to detect at least one hybrid individual — a threshold they term the "hybrid signal". They report a 100% correct detection rate across all scenarios when the fixation index (Fst) between the source populations was sufficient, though performance varied with sample size, number of markers used, and the generation of hybridisation.
The study also critiques the wide variation in how researchers interpret STRUCTURE output in published conservation studies, and proposes clearer guidelines for threshold-setting. The findings are relevant to researchers in conservation genetics, population genomics, and those applying model-based clustering methods to wildlife management questions. As a preprint, the results have not yet been peer-reviewed. The conclusions apply to the specific task of hybrid detection; broader claims about STRUCTURE's performance in other population-structure contexts are not addressed.
Sources
Read the original reporting — these are the public sources this summary draws from.
-
Primary sourcePreprint bioRxiv (Cold Spring Harbor Laboratory) · 2026-11-16The program STRUCTURE is good at detecting contemporary hybridization