ProtT-Affinity: Sequence-Based Protein-Protein Binding Affinity Prediction Using ProtT5 Embeddings

TL;DR

ProtT-Affinity is a sequence-based model using ProtT5 embeddings and a lightweight Transformer to predict protein-protein binding affinity, offering a practical alternative when structural data is unavailable.

Contribution

This work introduces ProtT-Affinity, a novel sequence-only approach combining ProtT5 embeddings with a lightweight Transformer for affinity prediction.

Findings

Achieves Pearson correlation of 0.628 and 0.459 on two test sets.

Competitive with widely used approaches despite not using structural data.

Large protein language models encode features relevant to binding energetics.

Abstract

Predicting the binding affinity of protein protein complexes directly from sequence remains a challenging problem, particularly in the absence of reliable structural information. Here I present ProtT Affinity, a sequence only model that combines ProtT5 embeddings with a lightweight Transformer architecture. The model is trained and evaluated on homology filtered subsets of the PDBBind database following a curation protocol consistent with prior structure based work. Across two independent test sets,ProtT Affinity reaches Pearson correlation coefficients of 0.628 and 0.459, respectively.Although its performance does not match the strongest structure based methods, it is competitive with several widely used approaches and provides a practical alternative when structural data are missing or uncertain. The results suggest that large protein language models capture features relevant to binding energetics, and that these features can be exploited to approximate affinity trends at scale.