Fine-tuning Protein Language Models with Deep Mutational Scanning improves Variant Effect Prediction
This work addresses the need for more accurate computational tools in genomics and clinical applications, though it is incremental as it builds on existing PLM methods with new training data.
The authors tackled the problem of improving protein variant effect prediction by fine-tuning Protein Language Models with experimental data from Deep Mutational Scanning, resulting in consistent performance gains on held-out test sets and benchmarks like ProteinGym and ClinVar.
Protein Language Models (PLMs) have emerged as performant and scalable tools for predicting the functional impact and clinical significance of protein-coding variants, but they still lag experimental accuracy. Here, we present a novel fine-tuning approach to improve the performance of PLMs with experimental maps of variant effects from Deep Mutational Scanning (DMS) assays using a Normalised Log-odds Ratio (NLR) head. We find consistent improvements in a held-out protein test set, and on independent DMS and clinical variant annotation benchmarks from ProteinGym and ClinVar. These findings demonstrate that DMS is a promising source of sequence diversity and supervised training data for improving the performance of PLMs for variant effect prediction.