Jonathan Terhorst

This paper introduces dreems, a graph neural network that efficiently estimates effective migration surfaces from genetic data, achieving lower error and faster computation than existing methods.

The paper develops a model of linkage disequilibrium evolution to show how genetic background and selection strength influence variant-level architecture across populations.

The paper introduces demestats, a software library to compute coalescence rates for demographic models, showing improved accuracy in inferring recent population history from genetic data.

The paper derives an alternative to the beta-model for genetic architecture under stabilizing selection, linking evolutionary theory to mixed models for improved prediction.

The paper introduces a new method for inferring complex demographic models from genetic variation data, achieving significant speedups over existing methods.

This paper shows that the DeltaK method for selecting K in ADMIXTURE can be inconsistent, often favoring K=2 even when more substructure exists.

This study uses ancient DNA to identify global patterns of natural selection across 7244 individuals from five regions, finding shared and region-specific adaptive signals.

This paper introduces a Bayesian method for inferring population size history from whole genome sequences, showing it is faster and more accurate than existing methods with full uncertainty quantification.

This study infers the geographic history of genetic ancestry using genomic tree sequences, revealing major population movements in Europe, Asia, and Africa.

The paper analyzes the statistical performance of demographic inference methods under continuous migration using theoretical bounds.

This paper analyzes the Li-Stephens haplotype copying model, deriving an algorithm to explore parameter space and demonstrating its impact on imputation and phasing tasks.

This paper introduces likelihood-based methods to detect natural selection, robust to population size fluctuations, using a probabilistic model of tree imbalance.

This paper introduces an exact method for SMC-based inference that avoids discretizing genealogies, improving speed and accuracy over existing approaches.

U-PASS provides a tool for power analysis and validity assessment of genetic association studies for binary traits, visualizing the trade-off between risk allele frequency and odds ratio.

The paper develops ASMC, a fast method to estimate coalescence times from SNP data, detects selection signals and finds enrichment of disease heritability in regions of background selection.

This paper develops a Gaussian process method to analyze genomic time series data from experimental evolution, enabling accurate detection and estimation of selection parameters.