Tyler C. Shimko

This paper introduces the EDEN family of foundation models for AI-programmable therapeutic design, demonstrating their ability to generate functional gene insertions, antibiotic peptides, and synthetic microbiomes.

DeCoDe optimizes degenerate codon libraries for protein-coding DNA, outperforming existing methods and supporting mixed-length libraries.

The paper introduces DFIM, a method to identify epistatic feature interactions in neural network models of regulatory DNA sequences, validated on simulated and in vivo data.