Wojtek Treyde
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Wojtek Treyde has a total of 287 co-authors. Here are the top co-authors by number of shared publications:
- Christopher J. Schofield
Chemistry Research Laboratory, Department of Chemistry and the Ineos Oxford Institute for Antimicrobial Research, University of Oxford, Oxford, UK
3 shared publications - Alexandre Tkatchenko
University of Luxembourg
2 shared publications - Alexis Dougha
Universite Paris Cite
2 shared publications - Anatol Ehrlich
University of Vienna
2 shared publications - Andre Schutzer Godoy
University of São Paulo
2 shared publications - Anthony Gitter
Department of Biostatistics and Medical Informatics, University of Wisconsin-Madison, Madison, WI, USA
2 shared publications
Latest Publications
- Open Access
- 17 November 2026
- Published by American Chemical Society (ACS) in ChemRxiv
CHELATE: High-precision structure prediction for metalloproteins
Accurate prediction of ligand binding modes in non-heme metalloproteins remains challenging because conventional docking methods often fail to correctly identify poses with accurate metal-coordination geometries. Here, we introduce CHELATE, a machine learning-based post-docking filter that predicts the root-mean-square deviation (RMSD) of ligand metal-binding groups (MBGs) to filter out inaccurate poses and prioritise those with native-like metal coordination. We curated a dataset of over 50,000 metalloprotein–ligand systems spanning more than 20 metal ions, split based on protein-ligand similarity to enable evaluation on structurally distinct systems. Cross-docking benchmarks showed that, although existing docking methods often sample poses with accurate metal-binding group placement, they fail to rank them correctly. We quantify this using the MBG success rate, defined as the fraction of top-ranked poses with an MBG RMSD ≤ 1 Å relative to the crystal structure. Applying CHELATE as a post-docking filter improved MBG success rates from a baseline of 18% across all systems to 32%, 34%, and 40% for AutoDock Vina, smina, and GNINA, respectively, for systems for which at least one docking pose passed the filter. Feature importance analysis indicates that CHELATE predictions are informed by chemically meaningful features associated with metal coordination. Representative case studies illustrate the scope and limitations of CHELATE. Its low computational cost and compatibility with existing docking workflows make it a promising filtering strategy for a broad range of structure prediction methods, including protein–ligand co-folding models.
- 17 November 2026
- Published by American Chemical Society (ACS) in ChemRxiv
Open AccessCHELATE: High-precision structure prediction for metalloproteins
Accurate prediction of ligand binding modes in non-heme metalloproteins remains challenging because conventional docking methods often fail to correctly identify poses with accurate metal-coordination geometries. Here, we introduce CHELATE, a machine learning-based post-docking filter that predicts the root-mean-square deviation (RMSD) of ligand metal-binding groups (MBGs) to filter out inaccurate poses and prioritise those with native-like metal coordination. We curated a dataset of over 50,000 metalloprotein–ligand systems spanning more than 20 metal ions, split based on protein-ligand similarity to enable evaluation on structurally distinct systems. Cross-docking benchmarks showed that, although existing docking methods often sample poses with accurate metal-binding group placement, they fail to rank them correctly. We quantify this using the MBG success rate, defined as the fraction of top-ranked poses with an MBG RMSD ≤ 1 Å relative to the crystal structure. Applying CHELATE as a post-docking filter improved MBG success rates from a baseline of 18% across all systems to 32%, 34%, and 40% for AutoDock Vina, smina, and GNINA, respectively, for systems for which at least one docking pose passed the filter. Feature importance analysis indicates that CHELATE predictions are informed by chemically meaningful features associated with metal coordination. Representative case studies illustrate the scope and limitations of CHELATE. Its low computational cost and compatibility with existing docking workflows make it a promising filtering strategy for a broad range of structure prediction methods, including protein–ligand co-folding models.
- Open Access
- 12 June 2026
- Published by Wiley in Chemistry – A European Journal
- Vol. 32 (36) , e71248
Substrate‐Selective Inhibition of the SARS‐CoV‐2 Papain‐Like Protease: Inhibition of Hydrolysis of Human Over Viral Substrates
The SARS‐CoV‐2 papain‐like protease (PL pro ) and the main protease (M pro ) catalyze hydrolysis of the viral polyproteins pp1a/1ab into functional nonstructural proteins. PL pro and M pro are medicinal chemistry targets, with M pro inhibitors being used for COVID‐19 treatment. PL pro also catalyzes hydrolysis of ubiquitin and interferon‐stimulated gene 15 (ISG15) from post‐translationally modified human proteins. Here we report how screening of reported deubiquitinase inhibitors using solid‐phase extraction coupled to mass spectrometry assays with oligopeptide substrates based on pp1a/1ab and on an ISG15‐modified human protein enabled the identification of substrate‐selective PL pro inhibitors. The results reveal that the deubiquitinase inhibitor ML364 selectively inhibits the deISGylase activity of isolated PL pro over its pp1a/1ab‐processing activity. Structure‐activity relationship and computational studies support the assignment of ML364 and derivatives as substrate‐selective PL pro inhibitors. The combined results provide proof‐of‐concept for developing substrate‐selective inhibitors of PL pro and, by implication, related proteolytic enzymes, including deubiquitinases.
- 12 June 2026
- Published by Wiley in Chemistry – A European Journal
- Vol. 32 (36) , e71248
Open AccessSubstrate‐Selective Inhibition of the SARS‐CoV‐2 Papain‐Like Protease: Inhibition of Hydrolysis of Human Over Viral Substrates
The SARS‐CoV‐2 papain‐like protease (PL pro ) and the main protease (M pro ) catalyze hydrolysis of the viral polyproteins pp1a/1ab into functional nonstructural proteins. PL pro and M pro are medicinal chemistry targets, with M pro inhibitors being used for COVID‐19 treatment. PL pro also catalyzes hydrolysis of ubiquitin and interferon‐stimulated gene 15 (ISG15) from post‐translationally modified human proteins. Here we report how screening of reported deubiquitinase inhibitors using solid‐phase extraction coupled to mass spectrometry assays with oligopeptide substrates based on pp1a/1ab and on an ISG15‐modified human protein enabled the identification of substrate‐selective PL pro inhibitors. The results reveal that the deubiquitinase inhibitor ML364 selectively inhibits the deISGylase activity of isolated PL pro over its pp1a/1ab‐processing activity. Structure‐activity relationship and computational studies support the assignment of ML364 and derivatives as substrate‐selective PL pro inhibitors. The combined results provide proof‐of‐concept for developing substrate‐selective inhibitors of PL pro and, by implication, related proteolytic enzymes, including deubiquitinases.
- Top Cited Papers
- 26 February 2026
- Published by American Chemical Society (ACS) in Journal of Chemical Information and Modeling
- Vol. 66 (6) , 3129-3149
A Computational Community Blind Challenge on Pan-Coronavirus Drug Discovery Data
Computational blind challenges offer critical, unbiased opportunities to assess and accelerate scientific progress, as demonstrated by a breadth of breakthroughs over the past decade. We report the outcomes and key insights from an open science community blind challenge focused on computational methods in drug discovery, using lead optimization data from the AI-driven Structure-enabled Antiviral Platform Discovery Consortium's pan-coronavirus antiviral discovery program, in partnership with Polaris and the OpenADMET project. This collaborative initiative invited global participants from both academia and industry to develop and apply computational methods to predict the biochemical potency and crystallographic ligand poses of small molecules against key coronavirus targets, Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) and Middle East Respiratory Syndrome Coronavirus (MERS-CoV) main protease (Mpro), as well as multiple ADMET assay end points, using previously undisclosed comprehensive experimental drug discovery data sets as benchmarks. By evaluating submissions across multiple tasks and compounds, we established performance leaderboards and conducted meta-analyses to assess methodological strengths, common pitfalls, and areas for improvement. This analysis provides a foundation for best practices in real-world machine learning evaluation, grounded in community-driven benchmarking. We also highlight how next-generation platforms, such as Polaris, enable rigorous challenge design, embedded evaluation frameworks, and broad community engagement. This paper reports the collective findings of the challenge, offering a high-level overview of the data, evaluation infrastructure, and top-performing strategies. We further provide context and support for the accompanying papers authored by the challenge participants in this special issue, which explore individual approaches in greater depth. Together, these contributions aim to advance reproducible, trustworthy, and high-impact computational methods in drug discovery, and to explore best practices and pitfalls in future blind challenge design and execution, including planned initiatives for the OpenADMET project.
- 26 February 2026
- Published by American Chemical Society (ACS) in Journal of Chemical Information and Modeling
- Vol. 66 (6) , 3129-3149
Top Cited PapersA Computational Community Blind Challenge on Pan-Coronavirus Drug Discovery Data
Computational blind challenges offer critical, unbiased opportunities to assess and accelerate scientific progress, as demonstrated by a breadth of breakthroughs over the past decade. We report the outcomes and key insights from an open science community blind challenge focused on computational methods in drug discovery, using lead optimization data from the AI-driven Structure-enabled Antiviral Platform Discovery Consortium's pan-coronavirus antiviral discovery program, in partnership with Polaris and the OpenADMET project. This collaborative initiative invited global participants from both academia and industry to develop and apply computational methods to predict the biochemical potency and crystallographic ligand poses of small molecules against key coronavirus targets, Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) and Middle East Respiratory Syndrome Coronavirus (MERS-CoV) main protease (Mpro), as well as multiple ADMET assay end points, using previously undisclosed comprehensive experimental drug discovery data sets as benchmarks. By evaluating submissions across multiple tasks and compounds, we established performance leaderboards and conducted meta-analyses to assess methodological strengths, common pitfalls, and areas for improvement. This analysis provides a foundation for best practices in real-world machine learning evaluation, grounded in community-driven benchmarking. We also highlight how next-generation platforms, such as Polaris, enable rigorous challenge design, embedded evaluation frameworks, and broad community engagement. This paper reports the collective findings of the challenge, offering a high-level overview of the data, evaluation infrastructure, and top-performing strategies. We further provide context and support for the accompanying papers authored by the challenge participants in this special issue, which explore individual approaches in greater depth. Together, these contributions aim to advance reproducible, trustworthy, and high-impact computational methods in drug discovery, and to explore best practices and pitfalls in future blind challenge design and execution, including planned initiatives for the OpenADMET project.
- Open Access
- 6 January 2026
- Published by American Chemical Society (ACS) in ChemRxiv
A Computational Community Blind Challenge on Pan-Coronavirus Drug Discovery Data
Computational blind challenges offer critical, unbiased assessment opportunities to assess and accelerate scientific progress, as demonstrated by a breadth of breakthroughs over the last decade. We report the outcomes and key insights from an open science community blind challenge focused on computational methods in drug discovery, using lead optimization data from the AI-driven Structure-enabled Antiviral Platform (ASAP) Discovery Consortium’s pan-coronavirus antiviral discovery program, in partnership with Polaris and the OpenADMET project. This collaborative initiative invited global participants from both academia and industry to develop and apply computational methods to predict the biochemical potency and crystallographic ligand poses of small molecules against key coronavirus targets, Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) and Middle East Respiratory Syndrome Coronavirus (MERS-CoV) main protease (Mpro), as well as multiple ADMET assay endpoints, using previously undisclosed comprehensive experimental drug discovery datasets as benchmarks. By evaluating submissions across multiple tasks and compounds, we established performance leaderboards and conducted meta-analyses to assess methodological strengths, common pitfalls, and areas for improvement. This analysis provides a foundation for best practices in real-world machine learning evaluation, grounded in community-driven benchmarking. We also highlight how next-generation platforms, such as Polaris, enable rigorous challenge design, embedded evaluation frameworks, and broad community engagement. This paper reports the collective findings of the challenge, offering a high-level overview of the data, evaluation infrastructure, and top- performing strategies. We further provide context and support for the accompanying papers authored by the challenge participants in this special issue, which explore individual approaches in greater depth. Together, these contributions aim to advance reproducible, trustworthy, and high-impact computational methods in drug discovery, and to explore best practices and pitfalls in future blind challenge design and execution, including planned initiatives for the OpenADMET project.
- 6 January 2026
- Published by American Chemical Society (ACS) in ChemRxiv
Open AccessA Computational Community Blind Challenge on Pan-Coronavirus Drug Discovery Data
Computational blind challenges offer critical, unbiased assessment opportunities to assess and accelerate scientific progress, as demonstrated by a breadth of breakthroughs over the last decade. We report the outcomes and key insights from an open science community blind challenge focused on computational methods in drug discovery, using lead optimization data from the AI-driven Structure-enabled Antiviral Platform (ASAP) Discovery Consortium’s pan-coronavirus antiviral discovery program, in partnership with Polaris and the OpenADMET project. This collaborative initiative invited global participants from both academia and industry to develop and apply computational methods to predict the biochemical potency and crystallographic ligand poses of small molecules against key coronavirus targets, Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) and Middle East Respiratory Syndrome Coronavirus (MERS-CoV) main protease (Mpro), as well as multiple ADMET assay endpoints, using previously undisclosed comprehensive experimental drug discovery datasets as benchmarks. By evaluating submissions across multiple tasks and compounds, we established performance leaderboards and conducted meta-analyses to assess methodological strengths, common pitfalls, and areas for improvement. This analysis provides a foundation for best practices in real-world machine learning evaluation, grounded in community-driven benchmarking. We also highlight how next-generation platforms, such as Polaris, enable rigorous challenge design, embedded evaluation frameworks, and broad community engagement. This paper reports the collective findings of the challenge, offering a high-level overview of the data, evaluation infrastructure, and top- performing strategies. We further provide context and support for the accompanying papers authored by the challenge participants in this special issue, which explore individual approaches in greater depth. Together, these contributions aim to advance reproducible, trustworthy, and high-impact computational methods in drug discovery, and to explore best practices and pitfalls in future blind challenge design and execution, including planned initiatives for the OpenADMET project.
- Open Access
- 27 October 2025
- Published by Royal Society of Chemistry (RSC) in RSC Medicinal Chemistry
- Vol. 16 (12) , 6351-6367
C6-Alkoxy substituted penicillins are potent non-covalently binding inhibitors of the SARS-CoV-2 main protease
Introduction of a C6-alkoxy group on the penicillin scaffold can alter the mechanism of SARS-CoV-2 main protease (M pro ) inhibition.
- 27 October 2025
- Published by Royal Society of Chemistry (RSC) in RSC Medicinal Chemistry
- Vol. 16 (12) , 6351-6367
Open AccessC6-Alkoxy substituted penicillins are potent non-covalently binding inhibitors of the SARS-CoV-2 main protease
Introduction of a C6-alkoxy group on the penicillin scaffold can alter the mechanism of SARS-CoV-2 main protease (M pro ) inhibition.
- 18 December 2024
- Published by Royal Society of Chemistry (RSC) in Faraday Discussions
- Vol. 256, 639-663
Discovering synthesis targets: general discussion
- 18 December 2024
- Published by Royal Society of Chemistry (RSC) in Faraday Discussions
- Vol. 256, 639-663
Discovering synthesis targets: general discussion
- Open Access
QuickBind: A Light-Weight And Interpretable Molecular Docking Model
Predicting a ligand's bound pose to a target protein is a key component of early-stage computational drug discovery. Recent developments in machine learning methods have focused on improving pose quality at the cost of model runtime. For high-throughput virtual screening applications, this exposes a capability gap that can be filled by moderately accurate but fast pose prediction. To this end, we developed QuickBind, a light-weight pose prediction algorithm. We assess QuickBind on widely used benchmarks and find that it provides an attractive trade-off between model accuracy and runtime. To facilitate virtual screening applications, we augment QuickBind with a binding affinity module and demonstrate its capabilities for multiple clinically-relevant drug targets. Finally, we investigate the mechanistic basis by which QuickBind makes predictions and find that it has learned key physicochemical properties of molecular docking, providing new insights into how machine learning models generate protein-ligand poses. By virtue of its simplicity, QuickBind can serve as both an effective virtual screening tool and a minimal test bed for exploring new model architectures and innovations. Model code and weights are available at https://github.com/aqlaboratory/QuickBind .
Open AccessQuickBind: A Light-Weight And Interpretable Molecular Docking Model
Predicting a ligand's bound pose to a target protein is a key component of early-stage computational drug discovery. Recent developments in machine learning methods have focused on improving pose quality at the cost of model runtime. For high-throughput virtual screening applications, this exposes a capability gap that can be filled by moderately accurate but fast pose prediction. To this end, we developed QuickBind, a light-weight pose prediction algorithm. We assess QuickBind on widely used benchmarks and find that it provides an attractive trade-off between model accuracy and runtime. To facilitate virtual screening applications, we augment QuickBind with a binding affinity module and demonstrate its capabilities for multiple clinically-relevant drug targets. Finally, we investigate the mechanistic basis by which QuickBind makes predictions and find that it has learned key physicochemical properties of molecular docking, providing new insights into how machine learning models generate protein-ligand poses. By virtue of its simplicity, QuickBind can serve as both an effective virtual screening tool and a minimal test bed for exploring new model architectures and innovations. Model code and weights are available at https://github.com/aqlaboratory/QuickBind .
- Open Access
- 29 November 2022
- Published by Royal Society of Chemistry (RSC) in RSC Advances
- Vol. 12 (53) , 34557-34564
Bond dissociation energies of X–H bonds in proteins
A complete dataset of bond dissociation energies of X–H bonds in amino acids is reported, calculated using state of the art DFT methods. It shall aid interpreting experiments or predicting reactivity via machine learning.
- 29 November 2022
- Published by Royal Society of Chemistry (RSC) in RSC Advances
- Vol. 12 (53) , 34557-34564
Open AccessBond dissociation energies of X–H bonds in proteins
A complete dataset of bond dissociation energies of X–H bonds in amino acids is reported, calculated using state of the art DFT methods. It shall aid interpreting experiments or predicting reactivity via machine learning.