Kevin Korfmann

Postdoctoral researcher at the University of Pennsylvania. I received my PhD (summa cum laude) from the Technical University of Munich, where I worked on deep learning approaches to population genetics. My research focuses on integrating biology and computation — leveraging the stochastic nature of evolution to train deep learning models, with work spanning coalescent theory, ancestral recombination graphs, and simulation-based inference. Previously at the University of Oregon and TU Munich, with research stays at Queen Mary University of London and Imperial College London.

Documented Projects

Projects with full ReadTheDocs documentation

seqtec

Docs

Sequencing technology reference. Educational materials covering modern sequencing methods and their applications.

fastcxt

Docs

Fast coalescent translation toolkit. Efficient implementations for coalescence x translation operations.

pysinger

Docs

A Python rewrite of SINGER for sampling and inference of genealogies with recombination (ARG sampling).

cxt

Docs

Coalescence x Translation. Framework and documentation for the coalescent translation paradigm.

ancify

Docs

A polarization package for population genetics. Provides tools for ancestral allele polarization and related analyses.

FLI

Docs

A Friendly Likelihood-based Inference Guide. Educational resource for understanding likelihood methods in population genetics.

watchgen

Docs

The Watchmaker's Guide to Population Genetics. A comprehensive guide connecting watchmaking analogies to population genetic concepts.

A fork catalog for stdpopsim providing additional population genetic simulation models and demographic histories.

A fork catalog for stdpopsim with alternative population genetic models and simulation configurations.

stdvoidsim

Docs

A fork catalog for stdpopsim with standard population genetic models for additional species.

Research Software

Software accompanying published research

GNNcoal

Graph neural networks for demographic inference from genealogical trees under the coalescent model.

Python

sleepy

C++ forward-in-time simulator for selective sweeps under the weak seed bank (dormancy) model.

C++

CanningsSimulator

Simulator for the paper "Determinants of rapid adaptation in species with large variance in offspring production".

Jupyter Notebook

fork-cataloges-paper

Paper and analysis: Three fork catalogs for stdpopsim (stdferdowsim, stdgrimmsim, stdvoidsim).

Python

cxt

Coalescence and Translation: A language model approach to population genetics. Framework for the coalescent translation paradigm.

Python

fastcxt

Fast coalescent translation toolkit for efficient coalescence x translation operations.

Python

temporal-balancing-selection

Ensemble ResNet method for detecting balancing selection using temporal genomic data.

Jupyter Notebook

genetic_map_rn6

Genetic map of Littrell et al. 2018 for the rn6 rat genome assembly.

Python