MolSpectra: Pre-training 3D Molecular Representation with Multi-modal Energy Spectra
This addresses the need for more accurate molecular representations in computational chemistry, offering a novel integration of quantum effects, though it is incremental in enhancing pre-training methods.
The paper tackles the problem of learning 3D molecular representations by incorporating quantum mechanical energy spectra, which existing methods overlook, and results in improved performance on molecular property prediction and dynamics modeling benchmarks.
Establishing the relationship between 3D structures and the energy states of molecular systems has proven to be a promising approach for learning 3D molecular representations. However, existing methods are limited to modeling the molecular energy states from classical mechanics. This limitation results in a significant oversight of quantum mechanical effects, such as quantized (discrete) energy level structures, which offer a more accurate estimation of molecular energy and can be experimentally measured through energy spectra. In this paper, we propose to utilize the energy spectra to enhance the pre-training of 3D molecular representations (MolSpectra), thereby infusing the knowledge of quantum mechanics into the molecular representations. Specifically, we propose SpecFormer, a multi-spectrum encoder for encoding molecular spectra via masked patch reconstruction. By further aligning outputs from the 3D encoder and spectrum encoder using a contrastive objective, we enhance the 3D encoder's understanding of molecules. Evaluations on public benchmarks reveal that our pre-trained representations surpass existing methods in predicting molecular properties and modeling dynamics.