Computational biology

Funke Group

Jan Funke’s Group develops computer-vision and machine-learning techniques that are designed to work hand in hand with human annotators, as well as explainable AI methods that reveal the decision processes of machine learning models. Specifically, his group is interested in the identification of structures of interest in large datasets. His group’s work contributed substantially to the field of connectomics, as well as to the segmentation and tracking of cells in live-cell imaging datasets. Furthermore, Jan Funke’s group developed methods to discover subtle patterns in biological datasets, which can reveal previously unknown phenotypical differences. To further increase the utility and interpretability of machine learning methods, his group also designs models that directly incorporate biophysical constraints and domain knowledge. So far, their models have been used to count fluorophores beyond the diffraction limit and to infer synaptic plasticity rules from behavioral measurements.

Group members