AlphaFold's Bayesian Roots in Probability Kinematics

TL;DR

This paper reinterprets AlphaFold's success in protein structure prediction as a form of probability kinematics, providing a Bayesian perspective that enhances understanding and guides future model development.

Contribution

It establishes a theoretical link between AlphaFold's potential and probability kinematics, introducing a synthetic model to illustrate this probabilistic foundation.

Findings

AlphaFold's potential can be understood as Jeffrey conditioning.

A synthetic angular random walk model demonstrates the PK framework.

This reinterpretation offers a principled basis for future probabilistic models.

Abstract

The seminal breakthrough of AlphaFold in protein structure prediction relied on a learned potential energy function parameterized by deep models, in contrast to its successors AlphaFold2 and AlphaFold3, which lack an explicit probabilistic interpretation. While AlphaFold's potential was originally justified by heuristic analogy to physical potentials of mean force, we show that it can instead be understood as a principled instance of probability kinematics (PK), also known as Jeffrey conditioning, a generalization of Bayesian updating. This reinterpretation reveals that AlphaFold is a generalized Bayesian model that explicitly defines a posterior distribution over structures, providing a deeper explanation of its success and a foundation for future model design. To demonstrate this framework with precision, we introduce a tractable synthetic model in which an angular random walk prior is updated with distance-based evidence via PK, directly mirroring AlphaFold's mechanism. This setting allows us to explore the probabilistic foundations of AlphaFold in a clear and interpretable way. Our work connects a landmark in protein structure prediction to a broader class of compositional deep generative models and points to new opportunities for principled probabilistic approaches.