Kernel-Elastic Autoencoder for Molecular Design

TL;DR

The Kernel-Elastic Autoencoder (KAE) is a transformer-based generative model that achieves high diversity and accuracy in molecular design, enabling conditional generation and outperforming existing models in various benchmarks.

Contribution

KAE introduces two novel loss functions and demonstrates state-of-the-art performance in molecular generation, reconstruction, and conditional design tasks.

Findings

KAE achieves near-perfect reconstruction and high diversity in molecule generation.

KAE outperforms previous models in constrained optimization and docking affinity predictions.

KAE enables conditional molecule generation based on binding affinity scores.

Abstract

We introduce the Kernel-Elastic Autoencoder (KAE), a self-supervised generative model based on the transformer architecture with enhanced performance for molecular design. KAE is formulated based on two novel loss functions: modified maximum mean discrepancy and weighted reconstruction. KAE addresses the long-standing challenge of achieving valid generation and accurate reconstruction at the same time. KAE achieves remarkable diversity in molecule generation while maintaining near-perfect reconstructions on the independent testing dataset, surpassing previous molecule-generating models. KAE enables conditional generation and allows for decoding based on beam search resulting in state-of-the-art performance in constrained optimizations. Furthermore, KAE can generate molecules conditional to favorable binding affinities in docking applications as confirmed by AutoDock Vina and Glide scores, outperforming all existing candidates from the training dataset. Beyond molecular design, we anticipate KAE could be applied to solve problems by generation in a wide range of applications.