AbTune: Layer-wise selective fine-tuning of protein language models for antibodies

Abstract

AbstractO_ST_ABSMotivationC_ST_ABSAntibodies play central roles in immune defense and are widely used as therapeutic agents. However, the high structural and sequence diversity of antigen-binding loops, combined with limited experimental data and weak co-evolutionary signals, makes it difficult to develop generalizable predictive models. ResultsWe investigate test-time fine-tuning strategies to improve protein language model (pLM) performance in low-data settings, with a focus on antibody-related tasks. Systematic evaluations show that carefully constrained fine-tuning improves performance while preserving generalization. In particular, depth-selective fine-tuning consistently outperforms full-depth fine-tuning, with optimal performance achieved when tuning 50-75% of model layers for medium- to small-sized pLMs. We introduce AbTune, a test-time fine-tuning framework leveraging this depth-controlled adaptation strategy. Across antibody structure prediction, mutation effect prediction, and binding affinity prediction, AbTune outperforms standard pLM baselines and task-specific predictors on most benchmarks. We further analyze representation shifts, sequence-dependent adaptation behavior, and overfitting indicators, showing that fine-tuning depth, duration, and perplexity jointly determine performance. Availabilityhttps://github.com/haddocking/AbTune