Biology subjects

Montero-Blay, A.

Publications and source records attributed to Montero-Blay, A..

2 recordsLinked to original sources

bioRxiv2026

Agent-driven Model Development for RNA 3D Structure Prediction

Large language model (LLM) agents have shown promise in driving scientific discovery, but their effectiveness in complex, real-world biological problems remains underexplored. We ask whether a general-purpose LLM agent can drive semi-autonomous development of a model for a genuinely hard biological problem, RNA 3D structure prediction. We designed a development loop where, under a fixed budget and with human supervision, the agent iteratively proposed, implemented, trained, and evaluated model changes. Over 297 iterations, the model evolved from a randomly-initialised baseline to QuickFold, an 8.9M-parameter folding trunk that matches the strongest open-source baselines (RhoFold+, NuFold) on lDDT and TM-score within noise on a held-out test set at a fraction of their inference cost. We frame this less as a new predictor than as a case study in feedback-driven, agent-led model development.

bioinformatics↗

bioRxiv2020

Comparison of CRISPR and marker based methods for the engineering of phage T7

With the recent rise in interest in using lytic bacteriophages as therapeutic agents, there is an urgent requirement to understand their fundamental biology to enable the engineering of their genomes. Current methods of phage engineering rely on homologous recombination, followed by a system of selection to identify recombinant phages. For bacteriophage T7, the host genes cmk or trx have been used as a selection mechanism along with both type I and II CRISPR systems to select against wild-type phage and enrich for the desired mutant. Here we systematically compare all three systems; we show that the use of marker-based selection is the most efficient method and we use this to generate multiple T7 tail fiber mutants. Furthermore, we found the type II CRISPR-Cas system is easier to use and generally more efficient than a type I system in the engineering of phage T7. These results provide a foundation for the future, more efficient engineering of bacteriophage T7.

molecular biology↗