First functional AI-generated genomes: 302 candidate phage genomes designed
BrianHie · x · 2026-09-19
A study in Science (King et al.), highlighted by Nature Biotechnology, reports the first functional AI-generated genomes:
- The team used genomic language models Evo 1 and Evo 2, pretrained on large DNA datasets including 2M+ bacteriophage genomes; generated sequences showed diversity beyond what is seen in nature.
- They picked lytic phage ΦX174 (5.4 kb, 11 genes) as a design template and supervised fine-tuned the models on 15,000 Microviridae sequences.
- Computational filtering on sequence quality, tropism specificity, and evolutionary diversity produced 302 candidate phage genomes.
More from Research
- ABC Robotics Project Fully Released With 400+ Hours of Sim Data, Accepted to CoRL 2026 — GuanyaShi · 2026-09-19
- AM-Bench: A Simulation Suite and Benchmark for Aerial Manipulation — GuanyaShi · 2026-09-19
- First benchmark to compare official Jev against open-source lookalikes in the works — airesearch12 · 2026-09-19
- Study finds scientific beauty and impact are surprisingly correlated — jacobkimmel · 2026-09-19
- Trolley problem: Jev-style API turns jina-reranker-v3.5 into a ruthless decision engine — gaganghotra_ · 2026-09-19
- Probe guidance steers continuous diffusion LMs with just 1-3% extra inference compute — itsbautistam · 2026-09-19