Jonathan M. Goodman
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Jonathan M. Goodman has a total of 391 co-authors. Here are the top co-authors by number of shared publications:
Ian Paterson
University Chemical Laboratory, Lensfield Road, Cambridge, CB2 1EW, UK
20 shared publications- Luis Simón
Universidad de Salamanca
11 shared publications María A. Silva
Unilever Centre for Molecular Science Informatics, Department of Chemistry, University of Cambridge, Lensfield Road, Cambridge CB2 1EW, UK
11 shared publications- Cesare Gennari
Università degli Studi di Milano
10 shared publications - Ching Ching Lam
University of California, Los Angeles
10 shared publications Silvina C. Pellegrinet
Instituto de Química Rosario (CONICET), Facultad de Ciencias Bioquímicas y Farmacéuticas, Universidad Nacional de Rosario, Suipacha 531, Rosario (2000), Argentina
10 shared publications
Latest Publications
- Open Access
- 14 October 2026
- Published by Royal Society of Chemistry (RSC) in Digital Discovery
- Vol. 5 (10) , 4180-4190
Evaluating 2D and 3D graph encoders for infrared-based organic structure verification
The practical task for IR spectroscopy in analytical chemistry is structure verification: given a spectrum, can predicted spectra rank the correct structure highly among many candidates?
- 14 October 2026
- Published by Royal Society of Chemistry (RSC) in Digital Discovery
- Vol. 5 (10) , 4180-4190
Open AccessEvaluating 2D and 3D graph encoders for infrared-based organic structure verification
The practical task for IR spectroscopy in analytical chemistry is structure verification: given a spectrum, can predicted spectra rank the correct structure highly among many candidates?
- Open Access
- 19 July 2026
- Published by American Chemical Society (ACS) in ChemRxiv
Hybrid QM/ML approaches for the prediction of regioselectivity in Minisci C–H functionalisation reactions
Regioselectivity models are of key importance to the implementation of C–H functionalisations in synthetic route planning. In structure- and fragment-based drug discovery (SBDD/FBDD), ligand-target affinity is enhanced by elaborating weak hits into potent leads along vectors dictated by the protein structure. Consequently, the ability to functionalise a heterocyclic hit in a site-selective, and atom-economic manner could enable rapid exploration of structure-activity relationships (SAR). Designed to support this process, machine learning (ML) models for C–H regioselectivity prediction often report impressive accuracy, but progress in the development of these models, as well as their deployability, can be hindered by inaccessible or partially disclosed data, misaligned validation objectives, or an incomplete understanding of their domain of applicability. Herein, we present a quantum mechanics/machine learning (QM/ML) approach to predict the regioselectivity of Minisci reactions derived from carefully curated and publicly available data. We analyse and quantify data biases before validating our models, alongside relevant baselines, on a new, unique, FBDD-inspired experimental test set performed with high-throughput experimentation (HTE). We demonstrate superior accuracy on this experimental set compared to previous and baseline approaches, and offer a realistic perspective on their prospective performance. To support future development, we made our model and HTE dataset freely available at github.com/samgeoesp/minisci_regio.
- 19 July 2026
- Published by American Chemical Society (ACS) in ChemRxiv
Open AccessHybrid QM/ML approaches for the prediction of regioselectivity in Minisci C–H functionalisation reactions
Regioselectivity models are of key importance to the implementation of C–H functionalisations in synthetic route planning. In structure- and fragment-based drug discovery (SBDD/FBDD), ligand-target affinity is enhanced by elaborating weak hits into potent leads along vectors dictated by the protein structure. Consequently, the ability to functionalise a heterocyclic hit in a site-selective, and atom-economic manner could enable rapid exploration of structure-activity relationships (SAR). Designed to support this process, machine learning (ML) models for C–H regioselectivity prediction often report impressive accuracy, but progress in the development of these models, as well as their deployability, can be hindered by inaccessible or partially disclosed data, misaligned validation objectives, or an incomplete understanding of their domain of applicability. Herein, we present a quantum mechanics/machine learning (QM/ML) approach to predict the regioselectivity of Minisci reactions derived from carefully curated and publicly available data. We analyse and quantify data biases before validating our models, alongside relevant baselines, on a new, unique, FBDD-inspired experimental test set performed with high-throughput experimentation (HTE). We demonstrate superior accuracy on this experimental set compared to previous and baseline approaches, and offer a realistic perspective on their prospective performance. To support future development, we made our model and HTE dataset freely available at github.com/samgeoesp/minisci_regio.
- Open Access
- 5 March 2026
- Published by Royal Society of Chemistry (RSC) in Chemical Science
- Vol. 17 (18) , 9138-9147
DP5 without DFT: uncertainty-calibrated graph neural net accelerates structure confirmation via NMR
DP5q calculates the probability that a molecule is correctly assigned to an NMR spectrum. Extensive testing shows the results are comparable in accuracy to DFT-based DP5 and dramatically faster.
- 5 March 2026
- Published by Royal Society of Chemistry (RSC) in Chemical Science
- Vol. 17 (18) , 9138-9147
Open AccessDP5 without DFT: uncertainty-calibrated graph neural net accelerates structure confirmation via NMR
DP5q calculates the probability that a molecule is correctly assigned to an NMR spectrum. Extensive testing shows the results are comparable in accuracy to DFT-based DP5 and dramatically faster.
- Top Cited PapersOpen Access
- 11 February 2026
- Published by Royal Society of Chemistry (RSC) in Digital Discovery
- Vol. 5 (3) , 1237-1251
Enhancing molecular structure elucidation with reasoning-capable LLMs
We introduce a novel workflow that integrates reasoning-capable language models (LLMs) with chemical analysis tools to enhance molecular structure determination using NMR spectroscopy.
- 11 February 2026
- Published by Royal Society of Chemistry (RSC) in Digital Discovery
- Vol. 5 (3) , 1237-1251
Top Cited PapersOpen AccessEnhancing molecular structure elucidation with reasoning-capable LLMs
We introduce a novel workflow that integrates reasoning-capable language models (LLMs) with chemical analysis tools to enhance molecular structure determination using NMR spectroscopy.
- Open Access
- 21 January 2026
- Published by Royal Society of Chemistry (RSC) in Chemical Science
- Vol. 17 (12) , 5953-5964
Oligomerisation and stereoselective polymerisation of alkenes and alkynes using pyridyl-based Al( iii ) catalysts
The putative Al( iii ) cationic catalysts [RAl(2-py′) 2 (µ-R)AlR] + (2-py′ = 6-substituted pyridyl, R = Me, i Bu) can polymerise terminal alkenes with a high degree of stereocontrol and cyclotrimerise terminal alkynes.
- 21 January 2026
- Published by Royal Society of Chemistry (RSC) in Chemical Science
- Vol. 17 (12) , 5953-5964
Open AccessOligomerisation and stereoselective polymerisation of alkenes and alkynes using pyridyl-based Al( iii ) catalysts
The putative Al( iii ) cationic catalysts [RAl(2-py′) 2 (µ-R)AlR] + (2-py′ = 6-substituted pyridyl, R = Me, i Bu) can polymerise terminal alkenes with a high degree of stereocontrol and cyclotrimerise terminal alkynes.
- Top Cited PapersOpen Access
- 7 January 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Uncertainty-aware machine learning to predict non-cancer human toxicity for the global chemicals market
Humans are exposed to thousands of chemicals, yet limited toxicity data hinder effective management of their impacts on human health. High-performing machine learning models hold potential for addressing this gap, but their uncharacterized prediction performance across the wider range of chemicals undermines confidence in their results. We develop uncertainty-aware models to predict reproductive/developmental and general non-cancer human toxicity effect doses. Our well-calibrated models provide uncertainty estimates aligned with observed prediction errors and chemical familiarity. We predict toxicity with 95% confidence intervals for >100,000 globally marketed chemicals and identify toxicity and uncertainty hotspots. These results can be applied to inform decisions aimed at reducing potential human health impacts and guide targeted data generation and modeling efforts to reduce prediction uncertainty. Here, we show that enhancing transparency in prediction uncertainty provides key insights for building confidence in toxicity predictions, supporting the sound integration of machine learning-based predictions in chemical assessments.
- 7 January 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Top Cited PapersOpen AccessUncertainty-aware machine learning to predict non-cancer human toxicity for the global chemicals market
Humans are exposed to thousands of chemicals, yet limited toxicity data hinder effective management of their impacts on human health. High-performing machine learning models hold potential for addressing this gap, but their uncharacterized prediction performance across the wider range of chemicals undermines confidence in their results. We develop uncertainty-aware models to predict reproductive/developmental and general non-cancer human toxicity effect doses. Our well-calibrated models provide uncertainty estimates aligned with observed prediction errors and chemical familiarity. We predict toxicity with 95% confidence intervals for >100,000 globally marketed chemicals and identify toxicity and uncertainty hotspots. These results can be applied to inform decisions aimed at reducing potential human health impacts and guide targeted data generation and modeling efforts to reduce prediction uncertainty. Here, we show that enhancing transparency in prediction uncertainty provides key insights for building confidence in toxicity predictions, supporting the sound integration of machine learning-based predictions in chemical assessments.
- Open Access
- 9 October 2025
- Published by Royal Society of Chemistry (RSC) in Chemical Science
- Vol. 16 (45) , 21590-21599
Towards automatically verifying chemical structures: the powerful combination of 1 H NMR and IR spectroscopy
Experimental 1 H NMR and IR spectra can be scored against calculated data to verify candidate molecules. We show that combining these techniques is significantly more powerful for automated structure verification than using either one individually.
- 9 October 2025
- Published by Royal Society of Chemistry (RSC) in Chemical Science
- Vol. 16 (45) , 21590-21599
Open AccessTowards automatically verifying chemical structures: the powerful combination of 1 H NMR and IR spectroscopy
Experimental 1 H NMR and IR spectra can be scored against calculated data to verify candidate molecules. We show that combining these techniques is significantly more powerful for automated structure verification than using either one individually.
- Open Access
- 12 August 2025
- Published by American Chemical Society (ACS) in ChemRxiv
Advancing Structure Elucidation with a Flexible Multi-Spectral AI Model
Validating chemical synthesis success requires confirming the desired product using various analytical techniques. While spectroscopic data collection is increasingly automated, interpreting results remains a major bottleneck often requiring expert input. With advances in laboratory automation and high-throughput synthesis, this challenge is expected to intensify. We introduce the MultiModalSpectralTransformer (MMST), a machine learning method that predicts chemical structures directly from diverse spectral data (NMR, IR, MS). Trained on 4 million simulated compounds, MMST achieves 72% and 80 % as top-1 and top-3 accuracy respectively. To address out-of-distribution challenges, we implemented an active learning improvement cycle that generates molecules in similar chemical spaces, enabling the model to adapt to chemical structures beyond its original training data. We demonstrate MMST's capabilities through comprehensive benchmarking across diverse molecular weight ranges and chemical spaces. Notably, despite training solely on simulated data, MMST demonstrates good performance with experimental spectra. This research represents a significant advancement in automated structure elucidation, offering a powerful and adaptable tool that bridges the gap between simulated and real-world data.
- 12 August 2025
- Published by American Chemical Society (ACS) in ChemRxiv
Open AccessAdvancing Structure Elucidation with a Flexible Multi-Spectral AI Model
Validating chemical synthesis success requires confirming the desired product using various analytical techniques. While spectroscopic data collection is increasingly automated, interpreting results remains a major bottleneck often requiring expert input. With advances in laboratory automation and high-throughput synthesis, this challenge is expected to intensify. We introduce the MultiModalSpectralTransformer (MMST), a machine learning method that predicts chemical structures directly from diverse spectral data (NMR, IR, MS). Trained on 4 million simulated compounds, MMST achieves 72% and 80 % as top-1 and top-3 accuracy respectively. To address out-of-distribution challenges, we implemented an active learning improvement cycle that generates molecules in similar chemical spaces, enabling the model to adapt to chemical structures beyond its original training data. We demonstrate MMST's capabilities through comprehensive benchmarking across diverse molecular weight ranges and chemical spaces. Notably, despite training solely on simulated data, MMST demonstrates good performance with experimental spectra. This research represents a significant advancement in automated structure elucidation, offering a powerful and adaptable tool that bridges the gap between simulated and real-world data.
- Open Access
- 6 May 2025
- Published by American Chemical Society (ACS) in ChemRxiv
Enhancing Molecular Structure Elucidation with Reasoning-Capable LLMs
We introduce a novel workflow integrating reasoning-capable language models with specialized chemical analysis tools to enhance molecular structure determination using nuclear magnetic resonance spectroscopy. Generally, structure elucidation involves generating candidate molecular structures, comparing their predicted spectral features to experimental data, and identifying the best-fitting structure. Our workflow systematically generates diverse molecular candidates through chemical synthesis predictions, regioisomer exploration, and direct spectral-based methods. The language model bridges the gap between quantitative data and chemical insight by evaluating candidates through a reasoning process that analyzes spectral evidence, explains discrepancies, and assesses overall structural plausibility, moving beyond simple numerical error. This LLM-driven reasoning stage proved crucial, increasing correct top-ranked structure identification accuracy by 26.4%. Simulated spectral data with introduced noise artifacts and solvent peaks further highlighted the robustness of our method, showing accuracy improvements by 35.3%. The language model's confidence scores effectively correlated with prediction accuracy, facilitating efficient triage of results. While currently focused on HSQC data, this framework offers a flexible foundation for next-generation structure elucidation tools combining chemical expertise with advanced reasoning capabilities.
- 6 May 2025
- Published by American Chemical Society (ACS) in ChemRxiv
Open AccessEnhancing Molecular Structure Elucidation with Reasoning-Capable LLMs
We introduce a novel workflow integrating reasoning-capable language models with specialized chemical analysis tools to enhance molecular structure determination using nuclear magnetic resonance spectroscopy. Generally, structure elucidation involves generating candidate molecular structures, comparing their predicted spectral features to experimental data, and identifying the best-fitting structure. Our workflow systematically generates diverse molecular candidates through chemical synthesis predictions, regioisomer exploration, and direct spectral-based methods. The language model bridges the gap between quantitative data and chemical insight by evaluating candidates through a reasoning process that analyzes spectral evidence, explains discrepancies, and assesses overall structural plausibility, moving beyond simple numerical error. This LLM-driven reasoning stage proved crucial, increasing correct top-ranked structure identification accuracy by 26.4%. Simulated spectral data with introduced noise artifacts and solvent peaks further highlighted the robustness of our method, showing accuracy improvements by 35.3%. The language model's confidence scores effectively correlated with prediction accuracy, facilitating efficient triage of results. While currently focused on HSQC data, this framework offers a flexible foundation for next-generation structure elucidation tools combining chemical expertise with advanced reasoning capabilities.
- Open Access
- 7 March 2025
- Published by Springer Nature in Research Square
Uncertainty-aware machine learning to predict non-cancer human toxicity for the global chemicals market
Humans are exposed to many chemicals, yet limited toxicity data hinder effectively managing their impact on human health. High-performing machine learning models hold potential for addressing this gap, but their uncharacterized prediction performance across the wider chemical space undermines confidence in their results. We developed uncertainty-aware models to predict reproductive/developmental and general non-cancer human toxicity points of departure, aimed at supporting chemical risk and impact assessments. Our well-calibrated models provide uncertainty estimates aligned with observed prediction errors and chemical familiarity. We predict toxicity with 95% confidence intervals for >130,000 marketed chemicals and identify toxicity and uncertainty hotspots, including organothiophosphates, PFAS, steroids and dioxin-like compounds. These insights help prioritize efforts to reduce human health risks and improve model performance. By enhancing transparency in prediction uncertainty, our work supports the sound integration of ML-based predictions in chemical assessments and provides key insights for systematically building confidence in digitally derived toxicity information.
- 7 March 2025
- Published by Springer Nature in Research Square
Open AccessUncertainty-aware machine learning to predict non-cancer human toxicity for the global chemicals market
Humans are exposed to many chemicals, yet limited toxicity data hinder effectively managing their impact on human health. High-performing machine learning models hold potential for addressing this gap, but their uncharacterized prediction performance across the wider chemical space undermines confidence in their results. We developed uncertainty-aware models to predict reproductive/developmental and general non-cancer human toxicity points of departure, aimed at supporting chemical risk and impact assessments. Our well-calibrated models provide uncertainty estimates aligned with observed prediction errors and chemical familiarity. We predict toxicity with 95% confidence intervals for >130,000 marketed chemicals and identify toxicity and uncertainty hotspots, including organothiophosphates, PFAS, steroids and dioxin-like compounds. These insights help prioritize efforts to reduce human health risks and improve model performance. By enhancing transparency in prediction uncertainty, our work supports the sound integration of ML-based predictions in chemical assessments and provides key insights for systematically building confidence in digitally derived toxicity information.