AlphaFold's Bayesian Roots in Probability Kinematics
This work provides a theoretical foundation for AlphaFold's success, offering a principled probabilistic interpretation that could guide future model design in protein structure prediction and beyond.
The authors reinterpret AlphaFold's learned potential as an instance of probability kinematics (Jeffrey conditioning), showing it is a generalized Bayesian model that defines a posterior over structures. They introduce a synthetic model to demonstrate this connection, linking AlphaFold to broader compositional deep generative models.
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.