Lukas Villiger

This paper introduces EVOLVEpro, a few-shot learning framework using protein language models to rapidly improve protein activity with minimal experimental data.

This paper introduces RADARS, a programmable RNA-sensing technology that uses ADAR-mediated RNA editing to control protein expression based on target RNA transcripts in mammalian cells, with applications in gene therapy and synthetic biology.

This paper develops BE-DICT, a deep learning model trained on high-throughput screens to predict base editing outcomes with high accuracy.

In vivo prime editing corrects phenylketonuria in mice by repairing the Pah gene mutation and reducing blood phenylalanine levels.