J. Terhorst

Identifiability and inference of phylogenetic birth-death models

The paper proves that phylogenetic birth-death models with piecewise polynomial rate functions are identifiable but shows rate function estimation remains challenging.

Direct detection of natural selection in Bronze Age Britain

The paper develops a method to estimate time-varying selection coefficients from ancient DNA data and identifies loci related to vitamin D and calcium in Bronze Age Britain.

Variational phylodynamic inference using pandemic-scale data

This paper introduces VBSKY, a method for analyzing large pathogen genetic datasets to estimate epidemiological parameters like reproduction number, validated against public health data.

A class of identifiable phylogenetic birth-death models

This paper shows that a class of phylogenetic birth-death models is identifiable, contrasting with previous results showing unidentifiability for other models.

Efficiently inferring the demographic history of many populations with allele count data

The paper introduces a method to efficiently compute the expected sample frequency spectrum for complex demographic histories described by general directed acyclic graphs, and applies it to estimate a basal Eurasian admixture event in humans.