A Unified Generative Framework for Scalable Chemical Reaction Network Exploration

Changsu Cao; Chenxi Hu; Dingshun Lv; Hung Q. Pham; Ji Chen; Jin Li; Kailai Lin; Weiluo Ren; Zechang Sun

arxiv: 2606.21002 · v1 · pith:7456REKInew · submitted 2026-06-19 · ⚛️ physics.chem-ph

A Unified Generative Framework for Scalable Chemical Reaction Network Exploration

Zechang Sun , Chenxi Hu , Kailai Lin , Jin Li , Changsu Cao , Dingshun Lv , Ji Chen , Weiluo Ren

show 1 more author

Hung Q. Pham

This is my paper

Pith reviewed 2026-06-26 13:12 UTC · model grok-4.3

classification ⚛️ physics.chem-ph

keywords chemical reaction networksgenerative modelingtransition statesrectified flowcomputational chemistryreaction enumerationnetwork exploration

0 comments

The pith

ByteCRN replaces iterative transition state searches with a generative rectified flow model to build chemical reaction networks at scale.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces ByteCRN as an end-to-end framework that pairs chemically informed reaction enumeration with a generative rectified flow model for producing and validating transition state structures. This unified approach targets the main bottlenecks in computational CRN exploration: the explosion of possible reactions and the high cost of finding and confirming reaction paths. A reader would care because reliable CRN maps are essential for mechanism discovery and synthesis planning, yet conventional methods become impractical for even moderately complex systems. The framework reports that it maintains accuracy on individual reactions while delivering large efficiency gains across entire networks.

Core claim

ByteCRN is an end-to-end framework for computational CRN exploration that combines chemically informed reaction enumeration with a generative rectified flow architecture for both transition state generation and reaction validation, where the model maps reactant-product pairs to candidate transition state structures and verifies connectivity by mapping back to reactants and products, thereby replacing the most expensive steps of conventional workflows.

What carries the argument

A generative rectified flow architecture that maps reactant-product pairs to candidate transition state structures and verifies connectivity by mapping back to reactants and products.

If this is right

ByteCRN achieves a 10-100-fold acceleration over traditional workflows.
It maintains high predictive fidelity for individual reactions.
At the network scale, it prunes approximately 70-90% of the enumerated reactions.
The framework identifies novel pathways in the cyanoacetaldehyde system and successfully models the gamma-ketohydroperoxide network.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

If the model generalizes well, the same generative replacement strategy could be applied to other costly steps such as conformational sampling in larger molecules.
The pruning step might allow CRN construction to scale to systems with thousands of species where exhaustive enumeration remains impossible.
Combining the generative outputs with experimental rate data could provide an iterative refinement loop that further improves network accuracy.

Load-bearing premise

The generative rectified flow model, trained on prior data, produces transition state structures that are sufficiently accurate and generalizable to replace iterative transition state searches and intrinsic reaction coordinate validations for new reactant-product pairs in the target chemical spaces.

What would settle it

Running a full traditional transition state search plus IRC validation on the same reactant-product pairs examined by ByteCRN and finding that the generative model misses or incorrectly classifies a substantial fraction of the kinetically relevant reactions.

read the original abstract

Chemical reaction networks (CRNs) are crucial for understanding reaction mechanisms and guiding chemical synthesis, yet the computational exploration remains limited by the combinatorial growth of chemical space, the reliability of reaction path screening, and the cost of evaluating thermodynamic and kinetic properties. Here, we present ByteCRN, an end-to-end framework for computational CRN exploration that combines chemically informed reaction enumeration with generative transition state modeling. A key component of our framework is a generative rectified flow architecture for both transition state generation and reaction validation, where it maps reactant-product pairs to candidate transition state structures and verifies connectivity by mapping back to reactants and products. This unified generative strategy replaces the most expensive steps of conventional computational workflows, namely iterative transition state search and intrinsic reaction coordinate validation, within a complete CRN construction pipeline. ByteCRN delivers a 10--100-fold acceleration over traditional workflows while maintaining high predictive fidelity for individual reactions. At the network scale, it effectively prunes $\sim$70-90% of the enumerated reactions, streamlining the exploration of complex reaction space. Its utility is illustrated through the discovery of novel pathways involving cyanoacetaldehyde and the successful modeling of the challenging $\gamma$-ketohydroperoxide network, demonstrating a practical, scalable approach to autonomous chemical exploration.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

ByteCRN puts a single rectified flow model in charge of both generating and validating transition states inside a full CRN enumeration pipeline, but the accuracy claims rest on thin evidence.

read the letter

The main takeaway is that this work builds a pipeline called ByteCRN that enumerates reactions in a chemically informed way, then uses one generative rectified flow model to produce candidate transition states from reactant-product pairs and check them by mapping back the other direction. That bidirectional step is meant to stand in for the usual iterative TS search plus IRC validation.

What is actually new is the decision to handle both generation and validation inside the same rectified flow architecture rather than bolting a generative model onto existing search tools. The paper shows the full loop applied to two concrete cases: new pathways in cyanoacetaldehyde and the gamma-ketohydroperoxide network. That end-to-end framing is a step beyond simply training a model on TS data.

The approach looks workable on paper for people who already run large CRN explorations and need something faster than standard DFT workflows. The reported 10-100x speedup and 70-90% pruning at network scale would matter if the model really keeps the right reactions.

The soft spot is exactly where the stress-test note flags it. The back-mapping only confirms that the model is internally consistent; it does not prove the generated saddle point lies on the true minimum-energy path or that valid reactions are not being dropped. No RMSD values, barrier errors, or success rates on held-out or out-of-distribution pairs appear in the abstract, and the training data details are missing. Without those checks, the pruning numbers could simply reflect whatever the model learned to like or dislike.

This paper is for computational chemists who build or use automated mechanism discovery tools. A reader already working on scalable CRN methods would get value from seeing how the pieces fit together, even if they end up running their own validation.

It deserves a serious referee. The combination is distinct enough and the target problem is real, so the methods and any quantitative benchmarks should be examined properly rather than desk-rejected.

Referee Report

2 major / 2 minor

Summary. The manuscript introduces ByteCRN, an end-to-end framework for chemical reaction network (CRN) exploration that integrates chemically informed reaction enumeration with a generative rectified flow model for transition state (TS) generation and bidirectional validation. The central claims are that this unified generative strategy replaces iterative TS searches and IRC validations, delivering 10-100-fold acceleration over traditional workflows while maintaining high predictive fidelity, and pruning ~70-90% of enumerated reactions at network scale, as illustrated on cyanoacetaldehyde and γ-ketohydroperoxide systems.

Significance. If the generative rectified flow model produces accurate TS geometries and energies for new reactant-product pairs, the framework could substantially accelerate CRN construction and enable exploration of larger chemical spaces. The bidirectional generative validation strategy represents a potentially impactful departure from conventional quantum chemistry workflows, provided the fidelity and generalizability claims are substantiated.

major comments (2)

[Validation Methodology] The validation section relies on bidirectional mapping (reactant-product to TS and back) to confirm connectivity, but this only establishes reconstruction consistency and does not demonstrate that generated saddle points correspond to true minimum-energy paths or that false negatives are avoided. Explicit out-of-distribution benchmarks (e.g., RMSD to reference TS structures, barrier height errors, or reaction success rates on held-out pairs from the cyanoacetaldehyde and γ-ketohydroperoxide spaces) are required to support the fidelity and pruning claims.
[Results, Network Exploration] The network-scale results claim 70-90% pruning of enumerated reactions while preserving fidelity, but without quantitative comparison to traditional TS search + IRC workflows on the same systems (including error bars, baseline success rates, and confirmation that pruned reactions are chemically invalid rather than model artifacts), the acceleration and pruning statistics cannot be assessed as load-bearing evidence.

minor comments (2)

[Abstract] The abstract states 'high predictive fidelity' and '10--100-fold acceleration' without accompanying quantitative metrics or references to specific tables/figures; adding these would improve clarity.
[Methods] Notation for the rectified flow model parameters and the bidirectional mapping functions should be defined explicitly in the methods section to avoid ambiguity when describing the generative process.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed and constructive report. We address each major comment below and indicate the revisions we will make to improve the manuscript.

read point-by-point responses

Referee: [Validation Methodology] The validation section relies on bidirectional mapping (reactant-product to TS and back) to confirm connectivity, but this only establishes reconstruction consistency and does not demonstrate that generated saddle points correspond to true minimum-energy paths or that false negatives are avoided. Explicit out-of-distribution benchmarks (e.g., RMSD to reference TS structures, barrier height errors, or reaction success rates on held-out pairs from the cyanoacetaldehyde and γ-ketohydroperoxide spaces) are required to support the fidelity and pruning claims.

Authors: We agree that bidirectional mapping alone primarily demonstrates reconstruction consistency rather than confirming true minimum-energy paths or quantifying false negatives. The current manuscript relies on this approach together with internal consistency checks for the reported fidelity. To strengthen the claims, we will add explicit out-of-distribution benchmarks in the revised validation section, including RMSD to reference TS geometries, barrier height errors relative to DFT calculations, and reaction success rates on held-out pairs from both the cyanoacetaldehyde and γ-ketohydroperoxide chemical spaces. These additions will better substantiate the generative model's ability to identify valid transition states. revision: yes
Referee: [Results, Network Exploration] The network-scale results claim 70-90% pruning of enumerated reactions while preserving fidelity, but without quantitative comparison to traditional TS search + IRC workflows on the same systems (including error bars, baseline success rates, and confirmation that pruned reactions are chemically invalid rather than model artifacts), the acceleration and pruning statistics cannot be assessed as load-bearing evidence.

Authors: We acknowledge that the network-scale results would be more convincing with direct head-to-head comparisons. The manuscript currently reports acceleration factors and pruning percentages relative to literature-reported costs of traditional workflows rather than identical-system benchmarks. In revision we will add a dedicated comparison subsection that includes wall-time and success-rate metrics against standard TS search + IRC protocols run on the same cyanoacetaldehyde and γ-ketohydroperoxide reaction sets, with error bars where multiple independent runs are feasible, and chemical analysis of a sample of pruned reactions to confirm they fail traditional validation. revision: yes

Circularity Check

0 steps flagged

No circularity; empirical model performance claims are independent of inputs

full rationale

The paper describes an end-to-end generative framework (ByteCRN) whose central results—acceleration factors, pruning percentages, and pathway discoveries—are presented as outcomes of applying a trained rectified-flow model to specific reaction networks (cyanoacetaldehyde, γ-ketohydroperoxide). No equations, derivations, or claims reduce by construction to fitted parameters renamed as predictions, self-citations that bear the load of uniqueness, or ansatzes smuggled via prior work. The bidirectional mapping is an architectural feature of the generative model, not a self-referential proof that forces the reported fidelity or pruning statistics. The derivation chain is therefore self-contained against external benchmarks and receives the default non-finding.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claims depend on the empirical performance of a trained generative model whose parameters are fitted to chemical data, plus the assumption that this model generalizes to the enumerated reactions in the target networks.

free parameters (1)

Rectified flow model parameters
Large number of weights in the generative architecture are fitted during training on chemical structure data.

axioms (1)

domain assumption The training data distribution sufficiently covers the chemical space of the target CRNs for reliable generalization to new reactions.
Required for the generative model to produce valid transition states outside the training set.

pith-pipeline@v0.9.1-grok · 5778 in / 1234 out tokens · 29996 ms · 2026-06-26T13:12:13.201861+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

61 extracted references · 3 linked inside Pith

[1]

Unsleber and Markus Reiher

Jan P. Unsleber and Markus Reiher. The Exploration of Chemical Reaction Networks.Annual Review of Physical Chemistry, 71(Volume 71, 2020):121–142, April 2020

2020
[2]

Blau, Matthew J

Mingjian Wen, Evan Walter Clark Spotte-Smith, Samuel M. Blau, Matthew J. McDermott, Aditi S. Krishnapriyan, and Kristin A. Persson. Chemical reaction networks and opportunities for machine learning.Nature Computational Science, 3(1):12–24, 2023

2023
[3]

Mechanism Deduction from Noisy Chemical Reaction Networks.J

Jonny Proppe and Markus Reiher. Mechanism Deduction from Noisy Chemical Reaction Networks.J. Chem. Theory Comput., 15(1):357–370, January 2019

2019
[4]

Ali Hussain Motagamwala and James A. Dumesic. Microkinetic Modeling: A Tool for Rational Catalyst Design. Chem. Rev., 121(2):1049–1076, January 2021

2021
[5]

Uncertainty-Aware First-Principles Exploration of Chemical Reaction Networks.J

Moritz Bensberg and Markus Reiher. Uncertainty-Aware First-Principles Exploration of Chemical Reaction Networks.J. Phys. Chem. A, 128(22):4532–4547, June 2024

2024
[6]

Vargas-Hernández, Ranga Rohit Seemakurthi, Pol Sanz Berman, Rodrigo García-Muelas, Alán Aspuru-Guzik, and Núria López

Santiago Morandi, Oliver Loveday, Tim Renningholtz, Sergio Pablo-García, Rodrigo A. Vargas-Hernández, Ranga Rohit Seemakurthi, Pol Sanz Berman, Rodrigo García-Muelas, Alán Aspuru-Guzik, and Núria López. An end-to-end framework for reactivity in heterogeneous catalysis.Nature Chemical Engineering, 3(3):169–180, 2026

2026
[7]

Ulissi, Andrew J

Zachary W. Ulissi, Andrew J. Medford, Thomas Bligaard, et al. To address surface reaction network complexity using scaling relations machine learning and dft calculations.Nature Communications, 8:14621, 2017

2017
[8]

Johnson, Mark J

Mengjie Liu, Alon Grinberg Dana, Matthew S. Johnson, Mark J. Goldman, Agnes Jocher, A. Mark Payne, Colin A. Grambow, Kehang Han, Nathan W. Yee, Emily J. Mazeau, Katrin Blondal, Richard H. West, C. Franklin Goldsmith, and William H. Green. Reaction Mechanism Generator v3.0: Advances in Automatic Mechanism Generation.J. Chem. Inf. Model., 61(6):2686–2696, June 2021

2021
[9]

Blau, Hetal D

Samuel M. Blau, Hetal D. Patel, Evan Walter Clark Spotte-Smith, Xiaowei Xie, Shyam Dwaraknath, and Kristin A. Persson. A chemically consistent graph architecture for massive reaction networks applied to solid-electrolyte interphase formation.Chemical Science, 12(13):4931–4939, 2021

2021
[10]

Autonomous Reaction Network Exploration in Homogeneous and Heterogeneous Catalysis.Top Catal, 65(1):6–39, February 2022

Miguel Steiner and Markus Reiher. Autonomous Reaction Network Exploration in Homogeneous and Heterogeneous Catalysis.Top Catal, 65(1):6–39, February 2022

2022
[11]

McDermott, Shyam S

Matthew J. McDermott, Shyam S. Dwaraknath, and Kristin A. Persson. A graph-based network for predicting chemical reaction pathways in solid-state materials synthesis.Nature Communications, 12:3097, 2021

2021
[12]

Szymanski, Bernardus Rendy, Yuxing Fei, Rishi E

Nathan J. Szymanski, Bernardus Rendy, Yuxing Fei, Rishi E. Kumar, Tanjin He, David Milsted, Matthew J. McDermott, Max Gallant, Ekin Dogus Cubuk, Amil Merchant, Haegyeom Kim, Anubhav Jain, Christopher J. Bartel, Kristin Persson, Yan Zeng, and Gerbrand Ceder. An autonomous laboratory for the accelerated synthesis of inorganic materials.Nature, 624(7990):86–...

2023
[13]

Leonov, Alexander J

Artem I. Leonov, Alexander J. S. Hammer, Slawomir Lach, S. Hessam M. Mehr, Dario Caramelli, Davide Angelone, Aamir Khan, Steven O’Sullivan, Matthew Craven, Liam Wilbraham, and Leroy Cronin. An integrated self-optimizing programmable chemical synthesis and reaction engine.Nat Commun, 15(1):1240, February 2024

2024
[14]

Autonomous Discovery of Unknown Reaction Pathways from Data by Chemical Reaction Neural Network.J

Weiqi Ji and Sili Deng. Autonomous Discovery of Unknown Reaction Pathways from Data by Chemical Reaction Neural Network.J. Phys. Chem. A, 125(4):1082–1092, February 2021

2021
[15]

A human-machine interface for automatic exploration of chemical reaction networks.Nature Communications, 15(1):3680, 2024

Miguel Steiner and Markus Reiher. A human-machine interface for automatic exploration of chemical reaction networks.Nature Communications, 15(1):3680, 2024

2024
[16]

Trends and Outlook of Computational Chemistry and Microkinetic Modeling for Catalytic Synthesis of Methanol and DME.Catalysts, 10(6):655, June 2020

Jongmin Park, Hyo Seok Kim, Won Bo Lee, and Myung-June Park. Trends and Outlook of Computational Chemistry and Microkinetic Modeling for Catalytic Synthesis of Methanol and DME.Catalysts, 10(6):655, June 2020

2020
[17]

Deepreactionnetworkexplorationataheterogeneous catalytic interface.Nat Commun, 13(1):4860, August 2022

QiyuanZhao, YinanXu, JeffreyGreeley, andBrettM.Savoie. Deepreactionnetworkexplorationataheterogeneous catalytic interface.Nat Commun, 13(1):4860, August 2022

2022
[18]

Chirality matching induces molecularity change and diastereodivergence in iridium–catalyzed allylation of propargylic c–h bonds.ACS Catalysis, 16(9):8846–8857, 2026

Ruiqi Ding, Pei-Pei Xie, Jin Zhu, Yi-Ming Wang, and Peng Liu. Chirality matching induces molecularity change and diastereodivergence in iridium–catalyzed allylation of propargylic c–h bonds.ACS Catalysis, 16(9):8846–8857, 2026. 12

2026
[19]

Young, Joseph J

Tom A. Young, Joseph J. Silcock, Alistair J. Sterling, and Fernanda Duarte. autode: Automated calculation of reaction energy profiles— application to organic and organometallic reactions.Angewandte Chemie International Edition, 60(8):4266–4274, 2021

2021
[20]

Grimmel, Miguel Steiner, Paul L

Alberto Baiardi, Stephanie A. Grimmel, Miguel Steiner, Paul L. Türtscher, Jan P. Unsleber, Thomas Weymuth, and Markus Reiher. Expansive Quantum Mechanical Exploration of Chemical Reaction Paths.Acc. Chem. Res., 55(1):35–43, January 2022

2022
[21]

Unsleber, Stephanie A

Jan P. Unsleber, Stephanie A. Grimmel, and Markus Reiher. Chemoton 2.0: Autonomous Exploration of Chemical Reaction Networks.J. Chem. Theory Comput., 18(9):5393–5409, September 2022

2022
[22]

Automated Exploration of Reaction Networks and Mechanisms Based on Metadynamics Nanoreactor Simulations.J

Yutai Zhang, Chao Xu, and Zhenggang Lan. Automated Exploration of Reaction Networks and Mechanisms Based on Metadynamics Nanoreactor Simulations.J. Chem. Theory Comput., 19(23):8718–8731, December 2023

2023
[23]

Efficient prediction of reaction paths through molecular graph and reaction network analysis.Chem

Yeonjoon Kim, Jin Woo Kim, Zeehyo Kim, and Woo Youn Kim. Efficient prediction of reaction paths through molecular graph and reaction network analysis.Chem. Sci., 9(4):825–835, January 2018

2018
[24]

Ogunfowora, Sanjay S

Qiyuan Zhao, Sai Mahit Vaddadi, Michael Woulfe, Lawal A. Ogunfowora, Sanjay S. Garimella, Olexandr Isayev, and Brett M. Savoie. Comprehensive exploration of graphically defined reaction spaces.Sci Data, 10(1):145, March 2023

2023
[25]

Simone Vernuccio and L. J. Broadbelt. Discerning complex reaction networks using automated generators.AIChE Journal, 65(8), 2019

2019
[26]

Gao, Joshua W

Connie W. Gao, Joshua W. Allen, William H. Green, and Richard H. West. Reaction mechanism generator: Automatic construction of chemical kinetic mechanisms.Computer Physics Communications, 203:212–225, 2016

2016
[27]

Simm and Markus Reiher

Gregor N. Simm and Markus Reiher. Context-driven exploration of complex chemical reaction networks.Journal of Chemical Theory and Computation, 13(12):6108–6119, 2017

2017
[28]

Coley, William H

Connor W. Coley, William H. Green, and Klavs F. Jensen. Rdchiral: An rdkit wrapper for handling stereochemistry in retrosynthetic template extraction and application.Journal of Chemical Information and Modeling, 59(6):2529– 2537, 2019

2019
[29]

Qiyuan Zhao and Brett M. Savoie. Simultaneously improving reaction coverage and computational cost in automated reaction prediction tasks.Nature Computational Science, 1(7):479–490, 2021

2021
[30]

González Laffitte, Klaus Weinbauer, Daniel Merkle, Jakob Lykke Andersen, Rolf Fagerberg, Thomas Gatter, and Peter F

Tieu-Long Phan, Marcos E. González Laffitte, Klaus Weinbauer, Daniel Merkle, Jakob Lykke Andersen, Rolf Fagerberg, Thomas Gatter, and Peter F. Stadler. Synkit: A graph-based python framework for rule-based reaction modeling and analysis.Journal of Chemical Information and Modeling, 65(24):13012–13019, 2025

2025
[31]

Joung, Michael H

Jae F. Joung, Michael H. Fong, Nicolo Casetti, et al. Electron flow matching for generative reaction mechanism prediction.Nature, 645:115–123, 2025

2025
[32]

The path of chemical reactions - the IRC approach.Acc

Kenichi Fukui. The path of chemical reactions - the IRC approach.Acc. Chem. Res., 14(12):363–368, December 1981

1981
[33]

Gonzalez and H

Carlos. Gonzalez and H. Bernhard. Schlegel. Reaction path following in mass-weighted internal coordinates.J. Phys. Chem., 94(14):5523–5527, July 1990

1990
[34]

Prediction of transition state structures of gas-phase chemical reactions via machine learning

Sunghwan Choi. Prediction of transition state structures of gas-phase chemical reactions via machine learning. Nat Commun, 14(1):1168, March 2023

2023
[35]

Neuralneb—neural networks can find reaction paths fast.Machine Learning: Science and Technology, 3(4):045022, 2022

Mathias Schreiner, Arghya Bhowmik, Tejs Vegge, Peter Bjørn Jørgensen, and Ole Winther. Neuralneb—neural networks can find reaction paths fast.Machine Learning: Science and Technology, 3(4):045022, 2022

2022
[36]

Transition1x - a dataset for building generalizable reactive machine learning potentials.Scientific Data, 9(779), 2022

Mathias Schreiner, Arghya Bhowmik, Tejs Vegge, Jonas Busk, and Ole Winther. Transition1x - a dataset for building generalizable reactive machine learning potentials.Scientific Data, 9(779), 2022

2022
[37]

Enoch C. Y. Yuan, Anmol Kumar, Xiang Guan, et al. Analyticalab initiohessian from a deep learning potential for transition state optimization.Nature Communications, 15:8865, 2024

2024
[38]

Schaaf, Chen Lin, Guangrun Wang, and Philip Torr

Kalyan Ramakrishnan, Lars L. Schaaf, Chen Lin, Guangrun Wang, and Philip Torr. Implicit neural representations for chemical reaction paths.J. Chem. Phys., 163(3):034109, July 2025. 13

2025
[39]

Harnessing machine learning to enhance transition state search with interatomic potentials and generative models.Advanced Science, 12(34):e06240, 2025

Qiyuan Zhao, Yunhong Han, Duo Zhang, Jiaxu Wang, Peichen Zhong, Taoyong Cui, Bangchen Yin, Yirui Cao, Haojun Jia, and Chenru Duan. Harnessing machine learning to enhance transition state search with interatomic potentials and generative models.Advanced Science, 12(34):e06240, 2025

2025
[40]

Chenru Duan, Yuanqi Du, Haojun Jia, and Heather J. Kulik. Accurate transition state generation with an object-aware equivariant elementary reaction diffusion model, 2023. arXiv:2304.06174

arXiv 2023
[41]

Diffusion-based generative ai for exploring transition states from 2d molecular graphs.Nature Communications, 15:341, 2024

Seonghwan Kim, Jeheon Woo, and Woo Youn Kim. Diffusion-based generative ai for exploring transition states from 2d molecular graphs.Nature Communications, 15:341, 2024

2024
[42]

React-ot: Optimal transport for generating transition state in chemical reactions, 2024

Chenru Duan, Guan-Horng Liu, Yuanqi Du, Tianrong Chen, Qiyuan Zhao, et al. React-ot: Optimal transport for generating transition state in chemical reactions, 2024. arXiv:2404.13430

arXiv 2024
[43]

Kovar, and Esther Heid

Leonard Galustian, Konstantin Mark, Johannes Karwounopoulos, Maximilian P.-P. Kovar, and Esther Heid. GoFlow: efficient transition state geometry prediction with flow matching and E(3)-equivariant neural networks. Digital Discovery, 4(12):3492–3501, December 2025

2025
[44]

Generative Model for Constructing Reaction Path from Initial to Final States.J

Akihide Hayashi, So Takamoto, Ju Li, Yuta Tsuboi, and Daisuke Okanohara. Generative Model for Constructing Reaction Path from Initial to Final States.J. Chem. Theory Comput., 21(3):1292–1305, February 2025

2025
[45]

Makoś, Niraj Verma, Eric C

Małgorzata Z. Makoś, Niraj Verma, Eric C. Larson, Marek Freindorf, and Elfi Kraka. Generative adversarial networks for transition state geometry prediction.The Journal of Chemical Physics, 155(2):024116, 2021

2021
[46]

Anstine, and Olexandr Isayev

Filipp Nikitin, Dylan M. Anstine, and Olexandr Isayev. Right into the saddle: Stereochemistry-aware generation of molecular transition states.ChemRxiv, 2026(0406), 2026

2026
[47]

Vandewiele, Heta A

Qiyuan Zhao, Nathan M. Vandewiele, Heta A. Patel, Abhishek Kumar, and Brett M. Savoie. Algorithmic explorations of unimolecular and bimolecular reaction spaces.Angewandte Chemie International Edition, 62(2):e202210693, 2023

2023
[48]

Barnes, David R

Emilio Martínez-Núñez, George L. Barnes, David R. Glowacki, Sabine Kopec, Daniel Peláez, Aurelio Rodríguez, Roberto Rodríguez-Fernández, Robin J. Shannon, James J. P. Stewart, Pablo G. Tahoces, and Saulo A. Vazquez. Automekin2021: An open-source program for automated reaction discovery.Journal of Computational Chemistry, 42(28):2036–2048, 2021

2036
[49]

Exploring paths of chemical transformations in molecular and periodic systems: An approach utilizing force.WIREs Computational Molecular Science, 11(6):e1538, 2021

Satoshi Maeda and Yu Harabuchi. Exploring paths of chemical transformations in molecular and periodic systems: An approach utilizing force.WIREs Computational Molecular Science, 11(6):e1538, 2021

2021
[50]

A new perspective on building efficient and expressive 3d equivariant graph neural networks.Advances in Neural Information Processing Systems, 36, 2023

Yuanqi Du, Limei Wang, Dieqiao Feng, Guifeng Wang, Shuiwang Ji, Carla P Gomes, Zhi-Ming Ma, et al. A new perspective on building efficient and expressive 3d equivariant graph neural networks.Advances in Neural Information Processing Systems, 36, 2023

2023
[51]

Kingma and Max Welling

Diederik P. Kingma and Max Welling. Auto-encoding variational bayes, 2013. arXiv:1312.6114

Pith/arXiv arXiv 2013
[52]

Diffusion-based generative AI for exploring transition states from 2D molecular graphs.Nat Commun, 15(1):341, January 2024

Seonghwan Kim, Jeheon Woo, and Woo Youn Kim. Diffusion-based generative AI for exploring transition states from 2D molecular graphs.Nat Commun, 15(1):341, January 2024

2024
[53]

Flow straight and fast: Learning to generate and transfer data with rectified flow, 2022

Xingchao Liu, Chengyue Gong, and Qiang Liu. Flow straight and fast: Learning to generate and transfer data with rectified flow, 2022. arXiv:2209.03003

Pith/arXiv arXiv 2022
[54]

Hermes, Khachik Sargsyan, Habib N

Eric D. Hermes, Khachik Sargsyan, Habib N. Najm, and Judit Zádor. Sella, an open-source automation-friendly molecular saddle point optimizer.Journal of Chemical Theory and Computation, 18(11):6974–6988, 2022

2022
[55]

Introducing gpu-acceleration into the python-based simulations of chemistry framework, 2024

Rui Li, Qiming Sun, Xing Zhang, and Garnet Kin-Lic Chan. Introducing gpu-acceleration into the python-based simulations of chemistry framework, 2024. arXiv:2407.09700

arXiv 2024
[56]

Enhancing gpu-acceleration in the python-based simulations of chemistry framework, 2024

Xiaojie Wu, Qiming Sun, Zhichen Pu, Tianze Zheng, Wenzhi Ma, Wen Yan, Xia Yu, Zhengxiao Wu, Mian Huo, Xiang Li, Weiluo Ren, Sheng Gong, Yumin Zhang, and Weihao Gao. Enhancing gpu-acceleration in the python-based simulations of chemistry framework, 2024. arXiv:2404.09452

arXiv 2024
[57]

Christoph Bannwarth, Sebastian Ehlert, and Stefan Grimme. GFN2-xTB—an accurate and broadly parametrized self-consistent tight-binding quantum chemical method with multipole electrostatics and density-dependent dispersion contributions.Journal of Chemical Theory and Computation, 15(3):1652–1671, 2019

2019
[58]

Uni-mol2: Exploring molecular pretraining model at scale, 2024

Xiaohong Ji, Zhen Wang, Zhifeng Gao, Hang Zheng, Linfeng Zhang, et al. Uni-mol2: Exploring molecular pretraining model at scale, 2024. arXiv:2406.14969. 14

arXiv 2024
[59]

UMAP: Uniform manifold approximation and projection for dimension reduction.arXiv e-prints, page arXiv:1802.03426, 2018

Leland McInnes, John Healy, and James Melville. UMAP: Uniform manifold approximation and projection for dimension reduction.arXiv e-prints, page arXiv:1802.03426, 2018

Pith/arXiv arXiv 2018
[60]

Wesołowski, and Felix Zeller

Philipp Pracht, Stefan Grimme, Christoph Bannwarth, Fabian Bohle, Sebastian Ehlert, Gereon Feldmann, Johannes Gorges, Marcel Müller, Tim Neudecker, Christoph Plett, Sebastian Spicher, Pit Steinbach, Patryk A. Wesołowski, and Felix Zeller. CREST—a program for the exploration of low-energy molecular chemical space. The Journal of Chemical Physics, 160(11):1...

2024
[61]

bottleneck

Colin A. Grambow, Lakshmi Pattanaik, and William H. Green. Reactants, products, and transition states of elementary chemical reactions based on quantum chemistry.Scientific Data, 7(137), 2020. 15 Appendix A Transition States as Ideal Latent Space of MEP In this study, modeling traditional TS searches and IRC validations offers practical advantages, includ...

2020

[1] [1]

Unsleber and Markus Reiher

Jan P. Unsleber and Markus Reiher. The Exploration of Chemical Reaction Networks.Annual Review of Physical Chemistry, 71(Volume 71, 2020):121–142, April 2020

2020

[2] [2]

Blau, Matthew J

Mingjian Wen, Evan Walter Clark Spotte-Smith, Samuel M. Blau, Matthew J. McDermott, Aditi S. Krishnapriyan, and Kristin A. Persson. Chemical reaction networks and opportunities for machine learning.Nature Computational Science, 3(1):12–24, 2023

2023

[3] [3]

Mechanism Deduction from Noisy Chemical Reaction Networks.J

Jonny Proppe and Markus Reiher. Mechanism Deduction from Noisy Chemical Reaction Networks.J. Chem. Theory Comput., 15(1):357–370, January 2019

2019

[4] [4]

Ali Hussain Motagamwala and James A. Dumesic. Microkinetic Modeling: A Tool for Rational Catalyst Design. Chem. Rev., 121(2):1049–1076, January 2021

2021

[5] [5]

Uncertainty-Aware First-Principles Exploration of Chemical Reaction Networks.J

Moritz Bensberg and Markus Reiher. Uncertainty-Aware First-Principles Exploration of Chemical Reaction Networks.J. Phys. Chem. A, 128(22):4532–4547, June 2024

2024

[6] [6]

Vargas-Hernández, Ranga Rohit Seemakurthi, Pol Sanz Berman, Rodrigo García-Muelas, Alán Aspuru-Guzik, and Núria López

Santiago Morandi, Oliver Loveday, Tim Renningholtz, Sergio Pablo-García, Rodrigo A. Vargas-Hernández, Ranga Rohit Seemakurthi, Pol Sanz Berman, Rodrigo García-Muelas, Alán Aspuru-Guzik, and Núria López. An end-to-end framework for reactivity in heterogeneous catalysis.Nature Chemical Engineering, 3(3):169–180, 2026

2026

[7] [7]

Ulissi, Andrew J

Zachary W. Ulissi, Andrew J. Medford, Thomas Bligaard, et al. To address surface reaction network complexity using scaling relations machine learning and dft calculations.Nature Communications, 8:14621, 2017

2017

[8] [8]

Johnson, Mark J

Mengjie Liu, Alon Grinberg Dana, Matthew S. Johnson, Mark J. Goldman, Agnes Jocher, A. Mark Payne, Colin A. Grambow, Kehang Han, Nathan W. Yee, Emily J. Mazeau, Katrin Blondal, Richard H. West, C. Franklin Goldsmith, and William H. Green. Reaction Mechanism Generator v3.0: Advances in Automatic Mechanism Generation.J. Chem. Inf. Model., 61(6):2686–2696, June 2021

2021

[9] [9]

Blau, Hetal D

Samuel M. Blau, Hetal D. Patel, Evan Walter Clark Spotte-Smith, Xiaowei Xie, Shyam Dwaraknath, and Kristin A. Persson. A chemically consistent graph architecture for massive reaction networks applied to solid-electrolyte interphase formation.Chemical Science, 12(13):4931–4939, 2021

2021

[10] [10]

Autonomous Reaction Network Exploration in Homogeneous and Heterogeneous Catalysis.Top Catal, 65(1):6–39, February 2022

Miguel Steiner and Markus Reiher. Autonomous Reaction Network Exploration in Homogeneous and Heterogeneous Catalysis.Top Catal, 65(1):6–39, February 2022

2022

[11] [11]

McDermott, Shyam S

Matthew J. McDermott, Shyam S. Dwaraknath, and Kristin A. Persson. A graph-based network for predicting chemical reaction pathways in solid-state materials synthesis.Nature Communications, 12:3097, 2021

2021

[12] [12]

Szymanski, Bernardus Rendy, Yuxing Fei, Rishi E

Nathan J. Szymanski, Bernardus Rendy, Yuxing Fei, Rishi E. Kumar, Tanjin He, David Milsted, Matthew J. McDermott, Max Gallant, Ekin Dogus Cubuk, Amil Merchant, Haegyeom Kim, Anubhav Jain, Christopher J. Bartel, Kristin Persson, Yan Zeng, and Gerbrand Ceder. An autonomous laboratory for the accelerated synthesis of inorganic materials.Nature, 624(7990):86–...

2023

[13] [13]

Leonov, Alexander J

Artem I. Leonov, Alexander J. S. Hammer, Slawomir Lach, S. Hessam M. Mehr, Dario Caramelli, Davide Angelone, Aamir Khan, Steven O’Sullivan, Matthew Craven, Liam Wilbraham, and Leroy Cronin. An integrated self-optimizing programmable chemical synthesis and reaction engine.Nat Commun, 15(1):1240, February 2024

2024

[14] [14]

Autonomous Discovery of Unknown Reaction Pathways from Data by Chemical Reaction Neural Network.J

Weiqi Ji and Sili Deng. Autonomous Discovery of Unknown Reaction Pathways from Data by Chemical Reaction Neural Network.J. Phys. Chem. A, 125(4):1082–1092, February 2021

2021

[15] [15]

A human-machine interface for automatic exploration of chemical reaction networks.Nature Communications, 15(1):3680, 2024

Miguel Steiner and Markus Reiher. A human-machine interface for automatic exploration of chemical reaction networks.Nature Communications, 15(1):3680, 2024

2024

[16] [16]

Trends and Outlook of Computational Chemistry and Microkinetic Modeling for Catalytic Synthesis of Methanol and DME.Catalysts, 10(6):655, June 2020

Jongmin Park, Hyo Seok Kim, Won Bo Lee, and Myung-June Park. Trends and Outlook of Computational Chemistry and Microkinetic Modeling for Catalytic Synthesis of Methanol and DME.Catalysts, 10(6):655, June 2020

2020

[17] [17]

Deepreactionnetworkexplorationataheterogeneous catalytic interface.Nat Commun, 13(1):4860, August 2022

QiyuanZhao, YinanXu, JeffreyGreeley, andBrettM.Savoie. Deepreactionnetworkexplorationataheterogeneous catalytic interface.Nat Commun, 13(1):4860, August 2022

2022

[18] [18]

Chirality matching induces molecularity change and diastereodivergence in iridium–catalyzed allylation of propargylic c–h bonds.ACS Catalysis, 16(9):8846–8857, 2026

Ruiqi Ding, Pei-Pei Xie, Jin Zhu, Yi-Ming Wang, and Peng Liu. Chirality matching induces molecularity change and diastereodivergence in iridium–catalyzed allylation of propargylic c–h bonds.ACS Catalysis, 16(9):8846–8857, 2026. 12

2026

[19] [19]

Young, Joseph J

Tom A. Young, Joseph J. Silcock, Alistair J. Sterling, and Fernanda Duarte. autode: Automated calculation of reaction energy profiles— application to organic and organometallic reactions.Angewandte Chemie International Edition, 60(8):4266–4274, 2021

2021

[20] [20]

Grimmel, Miguel Steiner, Paul L

Alberto Baiardi, Stephanie A. Grimmel, Miguel Steiner, Paul L. Türtscher, Jan P. Unsleber, Thomas Weymuth, and Markus Reiher. Expansive Quantum Mechanical Exploration of Chemical Reaction Paths.Acc. Chem. Res., 55(1):35–43, January 2022

2022

[21] [21]

Unsleber, Stephanie A

Jan P. Unsleber, Stephanie A. Grimmel, and Markus Reiher. Chemoton 2.0: Autonomous Exploration of Chemical Reaction Networks.J. Chem. Theory Comput., 18(9):5393–5409, September 2022

2022

[22] [22]

Automated Exploration of Reaction Networks and Mechanisms Based on Metadynamics Nanoreactor Simulations.J

Yutai Zhang, Chao Xu, and Zhenggang Lan. Automated Exploration of Reaction Networks and Mechanisms Based on Metadynamics Nanoreactor Simulations.J. Chem. Theory Comput., 19(23):8718–8731, December 2023

2023

[23] [23]

Efficient prediction of reaction paths through molecular graph and reaction network analysis.Chem

Yeonjoon Kim, Jin Woo Kim, Zeehyo Kim, and Woo Youn Kim. Efficient prediction of reaction paths through molecular graph and reaction network analysis.Chem. Sci., 9(4):825–835, January 2018

2018

[24] [24]

Ogunfowora, Sanjay S

Qiyuan Zhao, Sai Mahit Vaddadi, Michael Woulfe, Lawal A. Ogunfowora, Sanjay S. Garimella, Olexandr Isayev, and Brett M. Savoie. Comprehensive exploration of graphically defined reaction spaces.Sci Data, 10(1):145, March 2023

2023

[25] [25]

Simone Vernuccio and L. J. Broadbelt. Discerning complex reaction networks using automated generators.AIChE Journal, 65(8), 2019

2019

[26] [26]

Gao, Joshua W

Connie W. Gao, Joshua W. Allen, William H. Green, and Richard H. West. Reaction mechanism generator: Automatic construction of chemical kinetic mechanisms.Computer Physics Communications, 203:212–225, 2016

2016

[27] [27]

Simm and Markus Reiher

Gregor N. Simm and Markus Reiher. Context-driven exploration of complex chemical reaction networks.Journal of Chemical Theory and Computation, 13(12):6108–6119, 2017

2017

[28] [28]

Coley, William H

Connor W. Coley, William H. Green, and Klavs F. Jensen. Rdchiral: An rdkit wrapper for handling stereochemistry in retrosynthetic template extraction and application.Journal of Chemical Information and Modeling, 59(6):2529– 2537, 2019

2019

[29] [29]

Qiyuan Zhao and Brett M. Savoie. Simultaneously improving reaction coverage and computational cost in automated reaction prediction tasks.Nature Computational Science, 1(7):479–490, 2021

2021

[30] [30]

González Laffitte, Klaus Weinbauer, Daniel Merkle, Jakob Lykke Andersen, Rolf Fagerberg, Thomas Gatter, and Peter F

Tieu-Long Phan, Marcos E. González Laffitte, Klaus Weinbauer, Daniel Merkle, Jakob Lykke Andersen, Rolf Fagerberg, Thomas Gatter, and Peter F. Stadler. Synkit: A graph-based python framework for rule-based reaction modeling and analysis.Journal of Chemical Information and Modeling, 65(24):13012–13019, 2025

2025

[31] [31]

Joung, Michael H

Jae F. Joung, Michael H. Fong, Nicolo Casetti, et al. Electron flow matching for generative reaction mechanism prediction.Nature, 645:115–123, 2025

2025

[32] [32]

The path of chemical reactions - the IRC approach.Acc

Kenichi Fukui. The path of chemical reactions - the IRC approach.Acc. Chem. Res., 14(12):363–368, December 1981

1981

[33] [33]

Gonzalez and H

Carlos. Gonzalez and H. Bernhard. Schlegel. Reaction path following in mass-weighted internal coordinates.J. Phys. Chem., 94(14):5523–5527, July 1990

1990

[34] [34]

Prediction of transition state structures of gas-phase chemical reactions via machine learning

Sunghwan Choi. Prediction of transition state structures of gas-phase chemical reactions via machine learning. Nat Commun, 14(1):1168, March 2023

2023

[35] [35]

Neuralneb—neural networks can find reaction paths fast.Machine Learning: Science and Technology, 3(4):045022, 2022

Mathias Schreiner, Arghya Bhowmik, Tejs Vegge, Peter Bjørn Jørgensen, and Ole Winther. Neuralneb—neural networks can find reaction paths fast.Machine Learning: Science and Technology, 3(4):045022, 2022

2022

[36] [36]

Transition1x - a dataset for building generalizable reactive machine learning potentials.Scientific Data, 9(779), 2022

Mathias Schreiner, Arghya Bhowmik, Tejs Vegge, Jonas Busk, and Ole Winther. Transition1x - a dataset for building generalizable reactive machine learning potentials.Scientific Data, 9(779), 2022

2022

[37] [37]

Enoch C. Y. Yuan, Anmol Kumar, Xiang Guan, et al. Analyticalab initiohessian from a deep learning potential for transition state optimization.Nature Communications, 15:8865, 2024

2024

[38] [38]

Schaaf, Chen Lin, Guangrun Wang, and Philip Torr

Kalyan Ramakrishnan, Lars L. Schaaf, Chen Lin, Guangrun Wang, and Philip Torr. Implicit neural representations for chemical reaction paths.J. Chem. Phys., 163(3):034109, July 2025. 13

2025

[39] [39]

Harnessing machine learning to enhance transition state search with interatomic potentials and generative models.Advanced Science, 12(34):e06240, 2025

Qiyuan Zhao, Yunhong Han, Duo Zhang, Jiaxu Wang, Peichen Zhong, Taoyong Cui, Bangchen Yin, Yirui Cao, Haojun Jia, and Chenru Duan. Harnessing machine learning to enhance transition state search with interatomic potentials and generative models.Advanced Science, 12(34):e06240, 2025

2025

[40] [40]

Chenru Duan, Yuanqi Du, Haojun Jia, and Heather J. Kulik. Accurate transition state generation with an object-aware equivariant elementary reaction diffusion model, 2023. arXiv:2304.06174

arXiv 2023

[41] [41]

Diffusion-based generative ai for exploring transition states from 2d molecular graphs.Nature Communications, 15:341, 2024

Seonghwan Kim, Jeheon Woo, and Woo Youn Kim. Diffusion-based generative ai for exploring transition states from 2d molecular graphs.Nature Communications, 15:341, 2024

2024

[42] [42]

React-ot: Optimal transport for generating transition state in chemical reactions, 2024

Chenru Duan, Guan-Horng Liu, Yuanqi Du, Tianrong Chen, Qiyuan Zhao, et al. React-ot: Optimal transport for generating transition state in chemical reactions, 2024. arXiv:2404.13430

arXiv 2024

[43] [43]

Kovar, and Esther Heid

Leonard Galustian, Konstantin Mark, Johannes Karwounopoulos, Maximilian P.-P. Kovar, and Esther Heid. GoFlow: efficient transition state geometry prediction with flow matching and E(3)-equivariant neural networks. Digital Discovery, 4(12):3492–3501, December 2025

2025

[44] [44]

Generative Model for Constructing Reaction Path from Initial to Final States.J

Akihide Hayashi, So Takamoto, Ju Li, Yuta Tsuboi, and Daisuke Okanohara. Generative Model for Constructing Reaction Path from Initial to Final States.J. Chem. Theory Comput., 21(3):1292–1305, February 2025

2025

[45] [45]

Makoś, Niraj Verma, Eric C

Małgorzata Z. Makoś, Niraj Verma, Eric C. Larson, Marek Freindorf, and Elfi Kraka. Generative adversarial networks for transition state geometry prediction.The Journal of Chemical Physics, 155(2):024116, 2021

2021

[46] [46]

Anstine, and Olexandr Isayev

Filipp Nikitin, Dylan M. Anstine, and Olexandr Isayev. Right into the saddle: Stereochemistry-aware generation of molecular transition states.ChemRxiv, 2026(0406), 2026

2026

[47] [47]

Vandewiele, Heta A

Qiyuan Zhao, Nathan M. Vandewiele, Heta A. Patel, Abhishek Kumar, and Brett M. Savoie. Algorithmic explorations of unimolecular and bimolecular reaction spaces.Angewandte Chemie International Edition, 62(2):e202210693, 2023

2023

[48] [48]

Barnes, David R

Emilio Martínez-Núñez, George L. Barnes, David R. Glowacki, Sabine Kopec, Daniel Peláez, Aurelio Rodríguez, Roberto Rodríguez-Fernández, Robin J. Shannon, James J. P. Stewart, Pablo G. Tahoces, and Saulo A. Vazquez. Automekin2021: An open-source program for automated reaction discovery.Journal of Computational Chemistry, 42(28):2036–2048, 2021

2036

[49] [49]

Exploring paths of chemical transformations in molecular and periodic systems: An approach utilizing force.WIREs Computational Molecular Science, 11(6):e1538, 2021

Satoshi Maeda and Yu Harabuchi. Exploring paths of chemical transformations in molecular and periodic systems: An approach utilizing force.WIREs Computational Molecular Science, 11(6):e1538, 2021

2021

[50] [50]

A new perspective on building efficient and expressive 3d equivariant graph neural networks.Advances in Neural Information Processing Systems, 36, 2023

Yuanqi Du, Limei Wang, Dieqiao Feng, Guifeng Wang, Shuiwang Ji, Carla P Gomes, Zhi-Ming Ma, et al. A new perspective on building efficient and expressive 3d equivariant graph neural networks.Advances in Neural Information Processing Systems, 36, 2023

2023

[51] [51]

Kingma and Max Welling

Diederik P. Kingma and Max Welling. Auto-encoding variational bayes, 2013. arXiv:1312.6114

Pith/arXiv arXiv 2013

[52] [52]

Diffusion-based generative AI for exploring transition states from 2D molecular graphs.Nat Commun, 15(1):341, January 2024

Seonghwan Kim, Jeheon Woo, and Woo Youn Kim. Diffusion-based generative AI for exploring transition states from 2D molecular graphs.Nat Commun, 15(1):341, January 2024

2024

[53] [53]

Flow straight and fast: Learning to generate and transfer data with rectified flow, 2022

Xingchao Liu, Chengyue Gong, and Qiang Liu. Flow straight and fast: Learning to generate and transfer data with rectified flow, 2022. arXiv:2209.03003

Pith/arXiv arXiv 2022

[54] [54]

Hermes, Khachik Sargsyan, Habib N

Eric D. Hermes, Khachik Sargsyan, Habib N. Najm, and Judit Zádor. Sella, an open-source automation-friendly molecular saddle point optimizer.Journal of Chemical Theory and Computation, 18(11):6974–6988, 2022

2022

[55] [55]

Introducing gpu-acceleration into the python-based simulations of chemistry framework, 2024

Rui Li, Qiming Sun, Xing Zhang, and Garnet Kin-Lic Chan. Introducing gpu-acceleration into the python-based simulations of chemistry framework, 2024. arXiv:2407.09700

arXiv 2024

[56] [56]

Enhancing gpu-acceleration in the python-based simulations of chemistry framework, 2024

Xiaojie Wu, Qiming Sun, Zhichen Pu, Tianze Zheng, Wenzhi Ma, Wen Yan, Xia Yu, Zhengxiao Wu, Mian Huo, Xiang Li, Weiluo Ren, Sheng Gong, Yumin Zhang, and Weihao Gao. Enhancing gpu-acceleration in the python-based simulations of chemistry framework, 2024. arXiv:2404.09452

arXiv 2024

[57] [57]

Christoph Bannwarth, Sebastian Ehlert, and Stefan Grimme. GFN2-xTB—an accurate and broadly parametrized self-consistent tight-binding quantum chemical method with multipole electrostatics and density-dependent dispersion contributions.Journal of Chemical Theory and Computation, 15(3):1652–1671, 2019

2019

[58] [58]

Uni-mol2: Exploring molecular pretraining model at scale, 2024

Xiaohong Ji, Zhen Wang, Zhifeng Gao, Hang Zheng, Linfeng Zhang, et al. Uni-mol2: Exploring molecular pretraining model at scale, 2024. arXiv:2406.14969. 14

arXiv 2024

[59] [59]

UMAP: Uniform manifold approximation and projection for dimension reduction.arXiv e-prints, page arXiv:1802.03426, 2018

Leland McInnes, John Healy, and James Melville. UMAP: Uniform manifold approximation and projection for dimension reduction.arXiv e-prints, page arXiv:1802.03426, 2018

Pith/arXiv arXiv 2018

[60] [60]

Wesołowski, and Felix Zeller

Philipp Pracht, Stefan Grimme, Christoph Bannwarth, Fabian Bohle, Sebastian Ehlert, Gereon Feldmann, Johannes Gorges, Marcel Müller, Tim Neudecker, Christoph Plett, Sebastian Spicher, Pit Steinbach, Patryk A. Wesołowski, and Felix Zeller. CREST—a program for the exploration of low-energy molecular chemical space. The Journal of Chemical Physics, 160(11):1...

2024

[61] [61]

bottleneck

Colin A. Grambow, Lakshmi Pattanaik, and William H. Green. Reactants, products, and transition states of elementary chemical reactions based on quantum chemistry.Scientific Data, 7(137), 2020. 15 Appendix A Transition States as Ideal Latent Space of MEP In this study, modeling traditional TS searches and IRC validations offers practical advantages, includ...

2020