Molecular Conformation Generation via Shifting Scores
This work addresses a critical problem in computational chemistry for generating accurate 3D molecular conformations, with incremental improvements over existing diffusion-based methods.
The paper tackles molecular conformation generation by proposing a novel approach that models the disintegration of molecules as shifting force fields, leading to a distribution shift from Gaussian to Maxwell-Boltzmann for inter-atomic distances, and demonstrates advantages over state-of-the-art methods in experiments on molecular datasets.
Molecular conformation generation, a critical aspect of computational chemistry, involves producing the three-dimensional conformer geometry for a given molecule. Generating molecular conformation via diffusion requires learning to reverse a noising process. Diffusion on inter-atomic distances instead of conformation preserves SE(3)-equivalence and shows superior performance compared to alternative techniques, whereas related generative modelings are predominantly based upon heuristical assumptions. In response to this, we propose a novel molecular conformation generation approach driven by the observation that the disintegration of a molecule can be viewed as casting increasing force fields to its composing atoms, such that the distribution of the change of inter-atomic distance shifts from Gaussian to Maxwell-Boltzmann distribution. The corresponding generative modeling ensures a feasible inter-atomic distance geometry and exhibits time reversibility. Experimental results on molecular datasets demonstrate the advantages of the proposed shifting distribution compared to the state-of-the-art.