Machine learning approaches for the study of protein surfaces
![](https://www.epfl.ch/labs/lpdi/wp-content/uploads/2021/10/system-gecc346fa7_1920-1536x864.jpg)
Predicting interactions between proteins and other biomolecules based solely on structure remains a challenge in biology. Protein molecular surfaces display patterns of chemical and geometric features that fingerprint a protein’s modes of interactions with other biomolecules. Our lab hypothesized that proteins participating in similar interactions may share common fingerprints, independent of their evolutionary history. Fingerprints could be learned from large-scale datasets and we therefore developed MaSIF (Molecular Surface Interaction Fingerprinting), a conceptual framework based on a geometric deep learning method to capture fingerprints that are important for specific biomolecular interactions. MaSIF’s proof-of-concept has been demonstrated with three prediction challenges: protein pocket-ligand prediction, protein–protein interaction (PPI) site prediction and ultrafast scanning of protein surfaces for prediction of protein–protein complexes.
![](https://www.epfl.ch/labs/lpdi/wp-content/uploads/2021/07/masif-figure-904x1024.jpg)
Although MaSIF demonstrated reliable predictions three to four orders of magnitude faster than other state-of-the-art algorithms, the main limitations stem from its reliance on pre-computed meshes and handcrafted features, as well as significant computational time and memory requirements. We therefore developed dMaSIF (differentiable molecular surface interaction fingerprinting), a new architecture free of any pre-computed features. With this new tool in hand, all computations are performed on-the-fly, with minimal memory requirements.
![](https://www.epfl.ch/labs/lpdi/wp-content/uploads/2021/07/masif-figure2-1024x308.jpg)
References : Gainza P. et al Nature methods (2020) ; Sverrisson F. et al BioRxiv(2020)