Author

Gavin Ayres

University of Oxford

Listed by name: these papers carry no ORCID for this author, so different people with the same name may be mixed together.

Biology Disciplines

Click to filter author papers

Bioinformatics· 1 Genomics· 1

Signature Assays, Organisms & Methods

Click to filter · or open tag

Approach:deep learning· 2 Research Focus:benchmarking· 1 Assay:DNA methylation· 1 Research Focus:drug response / drug discovery· 1 Organism:human· 1 Assay:metagenomics· 1 Research Focus:protein design· 1 Assay:protein sequence· 1

Frequent Co-Authors

Geraldene MunsamyiD· 2 shared Noelia FerruziD· 2 shared Aaron W. KollaschiD· 1 shared Alexandros Papadopoulos· 1 shared Ali Taghibakhshi· 1 shared Alicia Frame· 1 shared Anne M. Bara· 1 shared Anthony Beardsworth CostaiD· 1 shared

Activity by Year

2024-2026 · peak 1/yr

Co-Author Collaboration Network

10 top collaborators · 45 mutual co-author links

Click any collaborator node to inspect or filter shared papers

Focal author (center)Selected co-authorORCID co-authorUnlinked name

Curved chords = mutual co-authorship

2 shared papersGenomics

Geraldene MunsamyORCID

Mutual co-authors in cluster

Author profile →

2 papers

Recent·Most cited

Designing AI-programmable therapeutics with the EDEN family of foundation models

Geraldene Munsamy, Gavin Ayres, Carla Greco, Keith Y K Kam, Gus Minto-Cowcher, John St John, Tanggis Bohnuud, Matthew H. Bakalar, William Chow, Robert J. Pecoraro, Marcelo D. T. Torres, Aaron W. Kollasch, Marcus H. Y. Leung, Hassan Sirelkhatim, Francesco Farina, , Philipp Lorenz

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.

bioRxiv · Feb 9, 2026 · PreprintGenomics

Conditional language models enable the efficient design of proficient enzymes

Geraldene Munsamy, R. Illanes, Silvia Fruncillo, Ioanna T. Nakou, Sebastian Lindner, Gavin Ayres, Lesley S. Sheehan, Steven J. Moss, Ulrich Eckhard, Philipp Lorenz, Noelia Ferruz

An enzyme-specific conditional language model generates active carbonic anhydrases and lactate dehydrogenases to user specification, some under 40% identity to natural proteins, with no retraining.

bioRxiv · May 3, 2024 · PreprintBioinformatics