Implicitly and Differentiably Representing Protein Surfaces and Interfaces

TL;DR

This paper proposes representing proteins as unions of signed distance functions (SDFs) for potential use in machine learning applications, demonstrating a novel approach inspired by recent advances in 3D geometry representation.

Contribution

It introduces a differentiable pipeline for protein surface representation using SDFs, bridging a gap between geometric modeling and biological applications.

Findings

Initial proof of concept shows feasibility

Potential for improved protein modeling in ML tasks

Lays groundwork for future validation experiments

Abstract

We introduce a pipeline for representing a protein, or protein complex, as the union of signed distance functions (SDFs) by representing each atom as a sphere with the appropriate radius. While this idea has been used previously as a way to render images of proteins, it has not, to our knowledge, been widely adopted in a machine learning setting. Mirroring recent successful work applying SDFs to represent 3D geometry, we present a proof of concept that this representation of proteins could be useful in several biologically relevant applications. We also propose further experiments that are necessary to validate the proposed approach.