Kristine Deibler

Reward-guided inference-time search over the Proteína-Complexa sequence-structure codesign model yields validated binders across 127 targets, kinases and a free carbohydrate, with codesigned sequences outperforming ProteinMPNN redesign.

The paper presents a model validation framework for molecular property prediction models in drug discovery, revealing four failure modes and showing that common evaluation protocols overestimate performance.

This paper introduces PeptideCLM-2, a chemical language model for therapeutic peptides, showing strong performance in predicting drug development endpoints.

This paper introduces a deep-learning model, PALM, that predicts peptide aggregation using protein language model embeddings, showing improved performance with larger datasets.

This paper presents computational methods for designing GPCR-binding miniproteins and validates them through high-throughput screening, producing agonists and antagonists for multiple GPCRs.