Generative Design of New-to-nature Biosynthetic Assembly Lines with Genomic Language Modeling
- 1Tatta Bio
Abstract
Reprogramming biosynthetic assembly lines can extend biosynthesis beyond the chemical space explored by nature. However, this remains difficult because assembly-line function depends on coordinated interactions across large multidomain enzymes. Here, we couple gLM2, a genomic language model trained on metagenomic sequences, with discrete diffusion and domain-level conditioning to enable generative design and optimization of biosynthetic gene clusters. We apply this approach to a chimeric type I polyketide synthase (PKS) engineered to produce {delta}-valerolactam, a molecule not naturally synthesized by PKSs. Through iterative redesign of two multi-domain regions in the context of the full PKS sequence, gLM2 progressively improved {delta}-valerolactam production, yielding variants with up to 9.4-fold higher titer than the starting enzyme. Together, these results demonstrate that evolutionary sequence information can be learned and applied to complex, multi-domain enzyme design problems, expanding biosynthetic assembly lines to produce molecules outside their natural biosynthetic repertoire.
Related
- Discovery of microbial intergenic features with genomic language modeling and multimodal search
- Generative Language Modeling for Antibody CDR Grafting and Alignment-driven De Novo Design
- Generative design of bacteriophages with genome language models
- Speciesformer learns conserved cellular states for cross-species generative virtual cell modeling
- Few-shot regulatory sequence design with a genomic language model
BibTeX
@misc{lanclos2026generative,
title = {{Generative Design of New-to-nature Biosynthetic Assembly Lines with Genomic Language Modeling}},
author = {N. Lanclos and K. Ibrahim and A. Cornman and M. Huang and V. Gill and A. Jiang and J. Abraham and J. Gin and Y. Chen and C. Petzold and J. Baerwald and T. Kortemme and J. Keasling and Y. Hwang},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.11.750945},
url = {https://www.biorxiv.org/content/10.64898/2026.09.11.750945v1},
note = {CC BY-NC},
}
RIS
TY - UNPB
TI - Generative Design of New-to-nature Biosynthetic Assembly Lines with Genomic Language Modeling
AU - N. Lanclos
AU - K. Ibrahim
AU - A. Cornman
AU - M. Huang
AU - V. Gill
AU - A. Jiang
AU - J. Abraham
AU - J. Gin
AU - Y. Chen
AU - C. Petzold
AU - J. Baerwald
AU - T. Kortemme
AU - J. Keasling
AU - Y. Hwang
PY - 2026
DA - 2026/09/11
PB - bioRxiv
DO - 10.64898/2026.09.11.750945
UR - https://www.biorxiv.org/content/10.64898/2026.09.11.750945v1
AB - Reprogramming biosynthetic assembly lines can extend biosynthesis beyond the chemical space explored by nature. However, this remains difficult because assembly-line function depends on coordinated interactions across large multidomain enzymes. Here, we couple gLM2, a genomic language model trained on metagenomic sequences, with discrete diffusion and domain-level conditioning to enable generative design and optimization of biosynthetic gene clusters. We apply this approach to a chimeric type I polyketide synthase (PKS) engineered to produce {delta}-valerolactam, a molecule not naturally synthesized by PKSs. Through iterative redesign of two multi-domain regions in the context of the full PKS sequence, gLM2 progressively improved {delta}-valerolactam production, yielding variants with up to 9.4-fold higher titer than the starting enzyme. Together, these results demonstrate that evolutionary sequence information can be learned and applied to complex, multi-domain enzyme design problems, expanding biosynthetic assembly lines to produce molecules outside their natural biosynthetic repertoire.
ER -