Oliver M. Turnbull

Generative Language Modeling for Antibody CDR Grafting and Alignment-driven De Novo Design

GenCDR uses LLaMa-based models to generate antibody CDRs with framework context, achieving high recovery and developable designs through alignment-driven training.

p-IgGen: A Paired Antibody Generative Language Model

p-IgGen generates paired heavy-light chain antibodies with biophysical properties similar to clinical-stage therapeutics.

The Observed T cell receptor Space database enables paired-chain repertoire mining, coherence analysis and language modelling

The paper introduces the OTS database, a curated resource of paired-chain TCR sequences, and develops a language model for analyzing TCR pairings and coherence.

Contextualising the developability risk of antibodies with lambda light chains using enhanced therapeutic antibody profiling

The paper updates a tool to assess developability risks of kappa and lambda antibodies using machine learning, finding lambda antibodies have higher risk but some are more tractable.

A Step Towards Generalisability: Training a Machine Learning Scoring Function for Structure-Based Virtual Screening

The paper develops a machine learning scoring function for virtual screening that avoids dataset biases and demonstrates its ability to identify important binding interactions through input attribution.