PETIMOT: A Novel Framework for Inferring Protein Motions from Sparse Data Using SE(3)-Equivariant Graph Neural Networks
This work addresses a fundamental open problem in structural biology by enabling more efficient and accurate inference of protein motions, which is crucial for understanding biological functions.
The paper tackles the problem of approximating protein conformational ensembles from sparse experimental data by introducing PETIMOT, a framework that uses SE(3)-equivariant graph neural networks and transfer learning from protein language models, achieving superior performance in time and accuracy compared to state-of-the-art methods.
Proteins move and deform to ensure their biological functions. Despite significant progress in protein structure prediction, approximating conformational ensembles at physiological conditions remains a fundamental open problem. This paper presents a novel perspective on the problem by directly targeting continuous compact representations of protein motions inferred from sparse experimental observations. We develop a task-specific loss function enforcing data symmetries, including scaling and permutation operations. Our method PETIMOT (Protein sEquence and sTructure-based Inference of MOTions) leverages transfer learning from pre-trained protein language models through an SE(3)-equivariant graph neural network. When trained and evaluated on the Protein Data Bank, PETIMOT shows superior performance in time and accuracy, capturing protein dynamics, particularly large/slow conformational changes, compared to state-of-the-art flow-matching approaches and traditional physics-based models.