DeepMech: A Machine Learning Framework for Chemical Reaction Mechanism Prediction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Das, Manajit, Hoque, Ajnabiul, Baranwal, Mayank, Sunoj, Raghavan B.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911311121612800
author Das, Manajit
Hoque, Ajnabiul
Baranwal, Mayank
Sunoj, Raghavan B.
author_facet Das, Manajit
Hoque, Ajnabiul
Baranwal, Mayank
Sunoj, Raghavan B.
contents Prediction of complete step-by-step chemical reaction mechanisms (CRMs) remains a major challenge. Whereas the traditional approaches in CRM tasks rely on expert-driven experiments or costly quantum chemical computations, contemporary deep learning (DL) alternatives ignore key intermediates and mechanistic steps and often suffer from hallucinations. We present DeepMech, an interpretable graph-based DL framework employing atom- and bond-level attention, guided by generalized templates of mechanistic operations (TMOps), to generate CRMs. Trained on our curated ReactMech dataset (~30K CRMs with 100K atom-mapped and mass-balanced elementary steps), DeepMech achieves 98.98+/-0.12% accuracy in predicting elementary steps and 95.94+/-0.21% in complete CRM tasks, besides maintaining high fidelity even in out-of-distribution scenarios as well as in predicting side and/or byproducts. Extension to multistep CRMs relevant to prebiotic chemistry, demonstrates the ability of DeepMech in effectively reconstructing 2 pathways from simple primordial substrates to complex biomolecules such as serine and aldopentose. Attention analysis identifies reactive atoms/bonds in line with chemical intuition, rendering our model interpretable and suitable for reaction design.
format Preprint
id arxiv_https___arxiv_org_abs_2509_15872
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DeepMech: A Machine Learning Framework for Chemical Reaction Mechanism Prediction
Das, Manajit
Hoque, Ajnabiul
Baranwal, Mayank
Sunoj, Raghavan B.
Chemical Physics
Artificial Intelligence
Machine Learning
Prediction of complete step-by-step chemical reaction mechanisms (CRMs) remains a major challenge. Whereas the traditional approaches in CRM tasks rely on expert-driven experiments or costly quantum chemical computations, contemporary deep learning (DL) alternatives ignore key intermediates and mechanistic steps and often suffer from hallucinations. We present DeepMech, an interpretable graph-based DL framework employing atom- and bond-level attention, guided by generalized templates of mechanistic operations (TMOps), to generate CRMs. Trained on our curated ReactMech dataset (~30K CRMs with 100K atom-mapped and mass-balanced elementary steps), DeepMech achieves 98.98+/-0.12% accuracy in predicting elementary steps and 95.94+/-0.21% in complete CRM tasks, besides maintaining high fidelity even in out-of-distribution scenarios as well as in predicting side and/or byproducts. Extension to multistep CRMs relevant to prebiotic chemistry, demonstrates the ability of DeepMech in effectively reconstructing 2 pathways from simple primordial substrates to complex biomolecules such as serine and aldopentose. Attention analysis identifies reactive atoms/bonds in line with chemical intuition, rendering our model interpretable and suitable for reaction design.
title DeepMech: A Machine Learning Framework for Chemical Reaction Mechanism Prediction
topic Chemical Physics
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2509.15872