Chemistry

Variational autoencoder-based chemical latent space for large molecular structures with 3D complexity

T. Ochiai, T. Inukai, et al.

Discover the groundbreaking NP-VAE, a novel deep-learning approach that excels in managing large molecular structures, particularly natural compounds with chirality. This innovative method not only achieves remarkable reconstruction accuracy but also enables the generation of novel compounds with enhanced functions. Research conducted by Toshiki Ochiai, Tensei Inukai, Manato Akiyama, Kairi Furui, Masahito Ohue, Nobuaki Matsumori, Shinsuke Inuki, Motonari Uesugi, Toshiaki Sunazuka, Kazuya Kikuchi, Hideaki Kakeya, and Yasubumi Sakakibara.

00:00

Playback language: English

Index

Abstract

This study introduces NP-VAE, a deep-learning method based on variational autoencoders, designed to manage large molecular structures, including natural compounds with chirality. NP-VAE constructs a chemical latent space from large compound datasets, achieving higher reconstruction accuracy and stable generative model performance. Exploration of this latent space facilitated comprehensive analysis of a compound library and generation of novel structures with optimized functions.

Publisher

Communications Chemistry

Published On

Nov 16, 2023

Authors

Toshiki Ochiai, Tensei Inukai, Manato Akiyama, Kairi Furui, Masahito Ohue, Nobuaki Matsumori, Shinsuke Inuki, Motonari Uesugi, Toshiaki Sunazuka, Kazuya Kikuchi, Hideaki Kakeya, Yasubumi Sakakibara

Related Publications

Explore these studies to deepen your understanding of the subject.

Engineering and Technology

Graphene-based 3D XNOR-VRRAM with ternary precision for neuromorphic computing

B. Alimkhanuly, J. Sohn, et al.

Chemistry

Representation of molecular structures with persistent homology for machine learning applications in chemistry

J. Townsend, C. P. Micucci, et al.

Chemistry

Leveraging molecular structure and bioactivity with chemical language models for de novo drug design

M. Moret, I. P. Angona, et al.

Chemistry

A generative artificial intelligence framework based on a molecular diffusion model for the design of metal-organic frameworks for carbon capture

H. Park, X. Yan, et al.

Listen, Learn & Level Up

Over 10,000 hours of research content in 25+ fields, available in 12+ languages.

No more digging through PDFs, just hit play and absorb the world's latest research in your language, on your time.

listen to research audio papers with researchbunny