Faster inference of complex demographic models from large allele frequency spectra
Enes Dilber, Jiatong Liang, Junyan Tan, Jonathan Terhorst · 2026
Read Faster inference of complex demographic models from large allele frequency spectra by Enes Dilber, Jiatong Liang, Junyan Tan, Jonathan Terhorst (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13472727.txt.
Description
Enes Dilber and co-authors present momi3, software for inferring population history from genetic variation data at scales that previously overwhelmed computation. Demographic inference reconstructs past population sizes, migrations, and mixtures from the spectrum of allele frequencies, but models with many populations or samples become extremely slow. Their 2026 paper rebuilds the method on a modern computing framework.
Implemented in Python with the JAX library, the tool supports continuous migration, graphics processor execution, automatic differentiation, standardised model input, and pruning of genealogical detail that no longer matters. Together these yield speedups of up to a thousandfold over existing methods. The authors demonstrate analyses of archaic admixture, the mixing of ancient human groups, using hundreds of human genomes, a scale they argue was impractical before. Code and documentation are publicly available.
Catalog record
| CREATOR | Enes Dilber, Jiatong Liang, Junyan Tan, Jonathan Terhorst |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 45 KB |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |