Multi-Type Point Cloud Autoencoder: A Complete Equivariant Embedding for Molecule Conformation and Pose
This work addresses the need for robust 3D molecular representations in computational chemistry, offering a novel method for handling molecular pose and conformation, though it is incremental in advancing autoencoder-based approaches.
The authors tackled the problem of representing molecular conformation and 3D orientation for tasks like modelling molecular dimers, clusters, or condensed phases, by developing Mo3ENet, an autoencoder that provides a rotatable, complete, and equivariant embedding, demonstrating its effectiveness on scalar and vector property prediction tasks.
Representations are a foundational component of any modelling protocol, including on molecules and molecular solids. For tasks that depend on knowledge of both molecular conformation and 3D orientation, such as the modelling of molecular dimers, clusters, or condensed phases, we desire a rotatable representation that is provably complete in the types and positions of atomic nuclei and roto-inversion equivariant with respect to the input point cloud. In this paper, we develop, train, and evaluate a new type of autoencoder, molecular O(3) encoding net (Mo3ENet), for multi-type point clouds, for which we propose a new reconstruction loss, capitalizing on a Gaussian mixture representation of the input and output point clouds. Mo3ENet is end-to-end equivariant, meaning the learned representation can be manipulated on O(3), a practical bonus. An appropriately trained Mo3ENet latent space comprises a universal embedding for scalar and vector molecule property prediction tasks, as well as other downstream tasks incorporating the 3D molecular pose, and we demonstrate its fitness on several such tasks.