Towards global reaction feasibility and robustness prediction with high throughput data and bayesian deep learning

Abstract Predicting organic reaction feasibility and robustness against environmental factors is challenging. We address this issue by integrating high throughput experimentation (HTE) and Bayesian deep learning. Diverging from existing HTE studies focused on niche chemical spaces, in this work, our...

Full description

Saved in:
Bibliographic Details
Main Authors: Haowen Zhong, Yilan Liu, Haibin Sun, Yuru Liu, Rentao Zhang, Baochen Li, Yi Yang, Yuqing Huang, Fei Yang, Frankie S. Mak, Klement Foo, Sen Lin, Tianshu Yu, Peng Wang, Xiaoxue Wang
Format: Article
Language:English
Published: Nature Portfolio 2025-05-01
Series:Nature Communications
Online Access:https://doi.org/10.1038/s41467-025-59812-0
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849309753290784768
author Haowen Zhong
Yilan Liu
Haibin Sun
Yuru Liu
Rentao Zhang
Baochen Li
Yi Yang
Yuqing Huang
Fei Yang
Frankie S. Mak
Klement Foo
Sen Lin
Tianshu Yu
Peng Wang
Xiaoxue Wang
author_facet Haowen Zhong
Yilan Liu
Haibin Sun
Yuru Liu
Rentao Zhang
Baochen Li
Yi Yang
Yuqing Huang
Fei Yang
Frankie S. Mak
Klement Foo
Sen Lin
Tianshu Yu
Peng Wang
Xiaoxue Wang
author_sort Haowen Zhong
collection DOAJ
description Abstract Predicting organic reaction feasibility and robustness against environmental factors is challenging. We address this issue by integrating high throughput experimentation (HTE) and Bayesian deep learning. Diverging from existing HTE studies focused on niche chemical spaces, in this work, our in-house HTE platform conducted 11,669 distinct acid amine coupling reactions in 156 working hours, yielding the most extensive single HTE dataset at a volumetric scale for industrial delivery. Our Bayesian neural network model achieved a benchmark for prediction accuracy of 89.48% for reaction feasibility. Furthermore, our fine-grained uncertainty disentanglement enables efficient active learning, reducing 80% of data requirements. Additionally, our uncertainty analysis effectively identifies out-of-domain reactions and evaluates reaction robustness or reproducibility against environmental factors for scaling up, offering a practical framework for navigating chemical spaces and designing highly robust industrial processes.
format Article
id doaj-art-bb707eeb374746f0b42bc918eb225644
institution Kabale University
issn 2041-1723
language English
publishDate 2025-05-01
publisher Nature Portfolio
record_format Article
series Nature Communications
spelling doaj-art-bb707eeb374746f0b42bc918eb2256442025-08-20T03:53:58ZengNature PortfolioNature Communications2041-17232025-05-0116111110.1038/s41467-025-59812-0Towards global reaction feasibility and robustness prediction with high throughput data and bayesian deep learningHaowen Zhong0Yilan Liu1Haibin Sun2Yuru Liu3Rentao Zhang4Baochen Li5Yi Yang6Yuqing Huang7Fei Yang8Frankie S. Mak9Klement Foo10Sen Lin11Tianshu Yu12Peng Wang13Xiaoxue Wang14ChemLexChemLexChemLexChemLexChemLexChemLexChemLexMegaRobo Technologies Co., Ltd.Zhejiang LaboratoryExperimental Drug Development Centre (EDDC), Agency for Science, Technology and Research (A*STAR)Experimental Drug Development Centre (EDDC), Agency for Science, Technology and Research (A*STAR)ChemLexSchool of Data Science, The Chinese University of Hong Kong - ShenzhenChemLexChemLexAbstract Predicting organic reaction feasibility and robustness against environmental factors is challenging. We address this issue by integrating high throughput experimentation (HTE) and Bayesian deep learning. Diverging from existing HTE studies focused on niche chemical spaces, in this work, our in-house HTE platform conducted 11,669 distinct acid amine coupling reactions in 156 working hours, yielding the most extensive single HTE dataset at a volumetric scale for industrial delivery. Our Bayesian neural network model achieved a benchmark for prediction accuracy of 89.48% for reaction feasibility. Furthermore, our fine-grained uncertainty disentanglement enables efficient active learning, reducing 80% of data requirements. Additionally, our uncertainty analysis effectively identifies out-of-domain reactions and evaluates reaction robustness or reproducibility against environmental factors for scaling up, offering a practical framework for navigating chemical spaces and designing highly robust industrial processes.https://doi.org/10.1038/s41467-025-59812-0
spellingShingle Haowen Zhong
Yilan Liu
Haibin Sun
Yuru Liu
Rentao Zhang
Baochen Li
Yi Yang
Yuqing Huang
Fei Yang
Frankie S. Mak
Klement Foo
Sen Lin
Tianshu Yu
Peng Wang
Xiaoxue Wang
Towards global reaction feasibility and robustness prediction with high throughput data and bayesian deep learning
Nature Communications
title Towards global reaction feasibility and robustness prediction with high throughput data and bayesian deep learning
title_full Towards global reaction feasibility and robustness prediction with high throughput data and bayesian deep learning
title_fullStr Towards global reaction feasibility and robustness prediction with high throughput data and bayesian deep learning
title_full_unstemmed Towards global reaction feasibility and robustness prediction with high throughput data and bayesian deep learning
title_short Towards global reaction feasibility and robustness prediction with high throughput data and bayesian deep learning
title_sort towards global reaction feasibility and robustness prediction with high throughput data and bayesian deep learning
url https://doi.org/10.1038/s41467-025-59812-0
work_keys_str_mv AT haowenzhong towardsglobalreactionfeasibilityandrobustnesspredictionwithhighthroughputdataandbayesiandeeplearning
AT yilanliu towardsglobalreactionfeasibilityandrobustnesspredictionwithhighthroughputdataandbayesiandeeplearning
AT haibinsun towardsglobalreactionfeasibilityandrobustnesspredictionwithhighthroughputdataandbayesiandeeplearning
AT yuruliu towardsglobalreactionfeasibilityandrobustnesspredictionwithhighthroughputdataandbayesiandeeplearning
AT rentaozhang towardsglobalreactionfeasibilityandrobustnesspredictionwithhighthroughputdataandbayesiandeeplearning
AT baochenli towardsglobalreactionfeasibilityandrobustnesspredictionwithhighthroughputdataandbayesiandeeplearning
AT yiyang towardsglobalreactionfeasibilityandrobustnesspredictionwithhighthroughputdataandbayesiandeeplearning
AT yuqinghuang towardsglobalreactionfeasibilityandrobustnesspredictionwithhighthroughputdataandbayesiandeeplearning
AT feiyang towardsglobalreactionfeasibilityandrobustnesspredictionwithhighthroughputdataandbayesiandeeplearning
AT frankiesmak towardsglobalreactionfeasibilityandrobustnesspredictionwithhighthroughputdataandbayesiandeeplearning
AT klementfoo towardsglobalreactionfeasibilityandrobustnesspredictionwithhighthroughputdataandbayesiandeeplearning
AT senlin towardsglobalreactionfeasibilityandrobustnesspredictionwithhighthroughputdataandbayesiandeeplearning
AT tianshuyu towardsglobalreactionfeasibilityandrobustnesspredictionwithhighthroughputdataandbayesiandeeplearning
AT pengwang towardsglobalreactionfeasibilityandrobustnesspredictionwithhighthroughputdataandbayesiandeeplearning
AT xiaoxuewang towardsglobalreactionfeasibilityandrobustnesspredictionwithhighthroughputdataandbayesiandeeplearning