Manuel Razo‐Mejia

This paper introduces a geometric deep learning method to model phenotype-to-fitness maps, showing it outperforms linear methods in capturing nonlinear structure and predicting fitness landscapes.

The paper identifies hundreds of unannotated translation products in E. coli using a high-throughput reverse genetic approach, including novel polypeptides and chimeric proteins.

The paper assesses fitness effects of 9,000 natural variants in yeast signaling pathways, finding that many non-neutral variants show signs of local adaptation.