eSMC: a statistical model to infer admixture events from individual genomics data

Area of Science:

• Population Genetics
• Genomics
• Evolutionary Biology

Background:

• Understanding species demographic history relies on inferring historical population admixture events.
• Current methods require multiple data sources, lacking approaches for single-source admixture inference.
• Pairwise Sequentially Markovian Coalescent (PSMC) estimates effective population size but does not infer admixture.

Purpose of the Study:

• To develop a novel method for inferring historical population admixture events from a single individual's genomic data.
• To extend the capabilities of existing population genetic models to address data limitations.

Main Methods:

• Proposed eSMC, an extended Pairwise Sequentially Markovian Coalescent (PSMC) model.
• Evaluated eSMC performance using simulated population admixture events with varying times and ratios.
• Applied eSMC to infer admixture events in human, donkey, and goat populations using real genomic data.

Main Results:

• eSMC demonstrated robust performance on both simulated and real genomic data.
• Simulated admixture events were accurately inferred across a range of times (5-100 kya) and ratios (1:1 to 4:1).
• Inferred admixture times for humans, donkeys, and goats aligned with historical domestication events.

Conclusions:

• eSMC effectively infers the timing of the most recent admixture event from single-individual genomic data.
• The developed method overcomes limitations of previous approaches, enabling new avenues for demographic history research.
• Source code for eSMC is publicly available for further research and application.