Lab Members

I like Evolution and Equations.

Stephan Peischl, PI

I use equations and computers to try to understand how living things evolve. My recent interests include: mutation load dynamics, evolution in regions of low recombination, local adaptation and chromsomal inversions, adaptation to spatially and / or temporally changing environments, and genome size evolution in microbes.

I leverage statistics and machine learning to study eco-evolutionary processes at a genomic resolution.

Karolina Wąchała, PhD Student

My research explores the eco-evolutionary dynamics of chromosomal inversions, with a focus on their role in facilitating local adaptation. I use individual-based simulations to investigate how inversions interact with, e.g., genomic architecture and population structure to shape patterns of genetic variation and differentiation. My work aims to identify the conditions under which inversions evolve and spread, and how they contribute to adaptive divergence between populations. Ultimately, I will develop a ML classifier for identifying targets of selection within inversions, and apply it to genomic data from vinegar flies and threespine sticklebacks, with the goal of uncovering novel insights into the adaptive significance of inversions in natural populations.

I am using statistical methods and machine learning algorithms to characterize the environmental factors shaping microbial community composition.

Ianis Vilela, PhD Student

Microbial communities and their environment are in a constant and fascinating interplay, shaping each other in complex ways. But how exactly does this happen? How can we characterize these dynamics? These are the kinds of questions that drove my curiosity during my Master's thesis and led me to discover methodological gaps in the field. In my PhD, I am developing a novel analysis tool designed to accurately characterize the environmental factors that shape microbial community compositions. This approach applies a set of statistical methods and machine learning algorithms, aiming to provide a comprehensive and versatile tool that can be applied across different settings and studies.

I study evolutionary principles, particularly the role of recombination in population genetics.

Stefan Strütt, PostDoc

I study evolutionary principles, particularly the role of recombination in population genetics. My research examines how recombination modifiers interact with evolutionary forces and shape genetic signatures.
During my Ph.D., I investigated the effects of transitions from outcrossing to selfing. Currently, I study how weak selection may confound sex-specific demographic inference in humans during their range expansion from the Middle East to Europe. Another focus is the role of inversions in promoting adaptation.
Through my work, I aim to contribute to a deeper understanding of the mechanisms and history of evolution.

I design interactive tools that make it easier to learn mathematical modeling in ecology and evolution.

Ana Hermina Ghenu, Project Leader

I connect theory with experiments in evolution and ecology to predict how bacteria respond to environmental stressors, like antibiotics and extreme heat. My future work will focus on bacterial recombination and mobile genetic elements, with the goal of using soil bacteria to understand prokaryotic speciation.

Thanks to funding from the UniBE Digitalization Strategy 2030 and Innovation office, I am leading a study to understand whether interactive web-apps can make students better at mathematical modelling in ecology and evolution.

Interested in our study? Click here.

I apply deep learning to map locally adapted SNPs inside chromosomal inversions.

Janosch Imhof, Master Student

Chromosomal inversions are common structural variants that enable local adaptation by suppressing recombination and preserving advantageous allele combinations. Understanding how these genomic signatures arise and how they can be identified is a core challenge in evolutionary and population genetics.

In a previous project I built an ML‑ready pipeline that converts simulated inversion‑cline data into feature‑vector tables for detecting adaptive alleles. My current master's thesis extends this work with deep learning: I treat inversion‑cline data as images that capture linkage‑disequilibrium patterns produced by suppressed recombination in heterokaryotypes. The aim of my project is to train a convolutional neural network to predict which SNPs within an inversion are locally adapted.

I investigate how purifying selection shapes genetic diversity in non-recombining regions.

Yannick Jauslin, Master Student

Using forward-in-time simulations under various distributions of fitness effects, I test the generalizability of the structured coalescent for purifying selection to biologically realistic conditions. My study will contribute to interpreting the genetic history of the Y chromosome and the expansion of humans through Europe during and after the Neolithic transition.