Publications
Featured Work
Dual-encoder contrastive learning accelerates enzyme discovery, 2026 [PNAS] ​[Supplementary Material] [Code​]
[App​]
Prior Select Work
Origin of life, biochemical evolution and networks:
The history of enzyme evolution embedded in metabolism, 2026 [bioRxiv]
Designing allostery-inspired response in mechanical networks, 2025 [PNAS]
Primitive purine biosynthesis connects ancient geochemistry to modern metabolism, 2024 [Nature Ecol. Evol.] [OpenAccess on bioRxiv]
Reaction networks resemble low-dimensional regular lattices, 2024 [J. Chem. Theory Comput.] [OpenAccess on ChemRxiv]
Protein cost minimization promotes the emergence of coenzyme redundancy, 2022 [PNAS]
Modern views of ancient metabolic networks, 2018 [Curr. Opin. Syst. Biol.]
Remnants of an ancient metabolism without Phosphate, 2017 [Cell]
Uncertainty of prebiotic scenarios: The case of the non-enzymatic Reverse Tricarboxylic Acid Cycle, 2015 [Scientific Reports]
Chemical simulation, synthetic biology, AI:
Ultra-high-throughput mapping of genetic design space, 2025 [Nature] [OpenAccess on bioRxiv]
Engineering synthetic phosphorylation signaling networks in human cells, 2025 [Science]
Analog physical systems can exhibit double descent, 2025 [arXiv]
Toward routine CSP of pharmaceuticals: A fully automated protocol using neural network potentials, 2025 [arXiv]
Computational electrosynthesis: Perspective on mechanistic questions, methodological approaches, and elucidating the role of the electrical double layer, 2025 [ J. Phys. Chem. C]
Enzyme substrate prediction from three-dimensional feature representations using space-filling curves, 2023 [J. Chem. Inf. Model.] [OpenAccess on bioRxiv]
Double descent demystified: Identifying, interpreting & ablating the sources of a deep learning puzzle, 2023 [arXiv]
Statistics and bias-free sampling of reaction mechanisms from reaction network models, 2023 [J. Phys. Chem. A] [OpenAccess on ChemRxiv]
Memorizing without overfitting: Bias, variance, and interpolation in overparameterized models, 2022 [Phys. Rev. Research]
Electron-passing neural networks for atomic charge prediction in systems with arbitrary molecular charge, 2020 [J. Chem. Inf. Model.] [OpenAccess on ChemRxiv]
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
TO REMOVE
Publications
​Featured Work
Dual-encoder contrastive learning accelerates enzyme discovery, 2026 [PNAS]
Selected Work
Science
The history of enzyme evolution embedded in metabolism, 2025 [bioRxiv]
Primitive purine biosynthesis connects ancient geochemistry to modern metabolism, 2024 [Nature Ecol. Evol]
Reaction networks resemble low-dimensional regular lattices, 2024 [J. Chem. Theory Comput]
Protein cost minimization promotes the emergence of coenzyme redundancy, 2022 [PNAS]
Modern views of ancient metabolic networks, 2018 [Curr. Opin. Syst. Biol]
Remnants of an Ancient Metabolism without Phosphate , 2017 [Cell]
AI-accelerated Tooling
Ultra-high-throughput mapping of genetic design space, 2026 [Nature]
Engineering synthetic phosphorylation signaling networks in human cells, 2025 [Science]
Toward routine CSP of pharmaceuticals: a fully automated protocol using neural network potentials, 2025 [arXiv]
Computational Electrosynthesis: A Perspective on Mechanistic Questions, Methodological Approaches, and Elucidating the Role of the Electrical Double Layer, 2025 [ J. Phys. Chem. C, 2025]
Enzyme substrate prediction from three-dimensional feature representations using space-filling curves, 2025 [J. Chem. Inf. Model]
Electron-Passing Neural Networks for Atomic Charge Prediction in Systems with Arbitrary Molecular Charge, 2020 [Chem. Inf. Model]
Methods & Models
Analog physical systems can exhibit double descent, 2025 [arXiv]
Designing allostery-inspired response in mechanical networks, 2025 [PNAS]
Double descent demystified: identifying, interpreting & ablating the sources of a deep learning puzzle, 2023 [arXiv]
Memorizing without overfitting: Bias, variance, and interpolation in over-parameterized models, 2020 [arXiv]
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Requests:
John
Computational Electrosynthesis: A Perspective on Mechanistic Questions, Methodological Approaches, and Elucidating the Role of the Electrical Double Layer — J. Phys. Chem. C
ECE vs DISP Mechanisms in Anodic Electrolysis of Benzyl Alcohols: Computational Prediction of Microscopic Rate Constants (not used)
Computational Electrosynthesis Study of Anodic Intramolecular Olefin Coupling: Elucidating the Role of the Electrical Double Layer
Derek
Electron-Passing Neural Networks for Atomic Charge Prediction in Systems with Arbitrary Molecular ChargeJ. Chem. Inf. Model, 2020
Toward routine CSP of pharmaceuticals: a fully automated protocol using neural network potentials — arXiv, 2025
Jason
Designing allostery-inspired response in mechanical networks — PNAS, 2025
Engineering synthetic phosphorylation signaling networks in human cells — Science, 2025
Memorizing without overfitting: Bias, variance, and interpolation in over-parameterized models — arXiv, 2020
Requests:
John
Computational Electrosynthesis: A Perspective on Mechanistic Questions, Methodological Approaches, and Elucidating the Role of the Electrical Double Layer — J. Phys. Chem. C
ECE vs DISP Mechanisms in Anodic Electrolysis of Benzyl Alcohols: Computational Prediction of Microscopic Rate Constants (not used)
Computational Electrosynthesis Study of Anodic Intramolecular Olefin Coupling: Elucidating the Role of the Electrical Double Layer
Derek
Electron-Passing Neural Networks for Atomic Charge Prediction in Systems with Arbitrary Molecular ChargeJ. Chem. Inf. Model, 2020
Toward routine CSP of pharmaceuticals: a fully automated protocol using neural network potentials — arXiv, 2025
Jason
Designing allostery-inspired response in mechanical networks — PNAS, 2025
Engineering synthetic phosphorylation signaling networks in human cells — Science, 2025
Memorizing without overfitting: Bias, variance, and interpolation in over-parameterized models — arXiv, 2020