1
|
Godinez WJ, Trifonov V, Fang B, Kuzu G, Pei L, Guiguemde WA, Martin EJ, King FJ, Jenkins JL, Skewes-Cox P. Compound Activity Prediction with Dose-Dependent Transcriptomic Profiles and Deep Learning. J Chem Inf Model 2024; 64:2695-2704. [PMID: 38293736 DOI: 10.1021/acs.jcim.3c01855] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 02/01/2024]
Abstract
Predicting compound activity in assays is a long-standing challenge in drug discovery. Computational models based on compound-induced gene expression signatures from a single profiling assay have shown promise toward predicting compound activity in other, seemingly unrelated, assays. Applications of such models include predicting mechanisms-of-action (MoA) for phenotypic hits, identifying off-target activities, and identifying polypharmacologies. Here, we introduce transcriptomics-to-activity transformer (TAT) models that leverage gene expression profiles observed over compound treatment at multiple concentrations to predict the compound activity in other biochemical or cellular assays. We built TAT models based on gene expression data from a RASL-seq assay to predict the activity of 2692 compounds in 262 dose-response assays. We obtained useful models for 51% of the assays, as determined through a realistic held-out set. Prospectively, we experimentally validated the activity predictions of a TAT model in a malaria inhibition assay. With a 63% hit rate, TAT successfully identified several submicromolar malaria inhibitors. Our results thus demonstrate the potential of transcriptomic responses over compound concentration and the TAT modeling framework as a cost-efficient way to identify the bioactivities of promising compounds across many assays.
Collapse
Affiliation(s)
- William J Godinez
- Novartis Institutes for BioMedical Research, Emeryville, California 94608, United States
| | - Vladimir Trifonov
- Novartis Institutes for BioMedical Research, San Diego, California 92121, United States
| | - Bin Fang
- Novartis Institutes for BioMedical Research, San Diego, California 92121, United States
| | - Guray Kuzu
- Novartis Institutes for BioMedical Research, San Diego, California 92121, United States
| | - Luying Pei
- Novartis Institutes for BioMedical Research, Emeryville, California 94608, United States
| | - W Armand Guiguemde
- Novartis Institutes for BioMedical Research, Emeryville, California 94608, United States
| | - Eric J Martin
- Novartis Institutes for BioMedical Research, Emeryville, California 94608, United States
| | - Frederick J King
- Novartis Institutes for BioMedical Research, San Diego, California 92121, United States
| | - Jeremy L Jenkins
- Novartis Institutes for BioMedical Research, Cambridge, Massachusetts 02139, United States
| | - Peter Skewes-Cox
- Novartis Institutes for BioMedical Research, Emeryville, California 94608, United States
| |
Collapse
|
2
|
Mullowney MW, Duncan KR, Elsayed SS, Garg N, van der Hooft JJJ, Martin NI, Meijer D, Terlouw BR, Biermann F, Blin K, Durairaj J, Gorostiola González M, Helfrich EJN, Huber F, Leopold-Messer S, Rajan K, de Rond T, van Santen JA, Sorokina M, Balunas MJ, Beniddir MA, van Bergeijk DA, Carroll LM, Clark CM, Clevert DA, Dejong CA, Du C, Ferrinho S, Grisoni F, Hofstetter A, Jespers W, Kalinina OV, Kautsar SA, Kim H, Leao TF, Masschelein J, Rees ER, Reher R, Reker D, Schwaller P, Segler M, Skinnider MA, Walker AS, Willighagen EL, Zdrazil B, Ziemert N, Goss RJM, Guyomard P, Volkamer A, Gerwick WH, Kim HU, Müller R, van Wezel GP, van Westen GJP, Hirsch AKH, Linington RG, Robinson SL, Medema MH. Artificial intelligence for natural product drug discovery. Nat Rev Drug Discov 2023; 22:895-916. [PMID: 37697042 DOI: 10.1038/s41573-023-00774-7] [Citation(s) in RCA: 16] [Impact Index Per Article: 16.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Accepted: 07/20/2023] [Indexed: 09/13/2023]
Abstract
Developments in computational omics technologies have provided new means to access the hidden diversity of natural products, unearthing new potential for drug discovery. In parallel, artificial intelligence approaches such as machine learning have led to exciting developments in the computational drug design field, facilitating biological activity prediction and de novo drug design for molecular targets of interest. Here, we describe current and future synergies between these developments to effectively identify drug candidates from the plethora of molecules produced by nature. We also discuss how to address key challenges in realizing the potential of these synergies, such as the need for high-quality datasets to train deep learning algorithms and appropriate strategies for algorithm validation.
Collapse
Affiliation(s)
| | - Katherine R Duncan
- Strathclyde Institute of Pharmacy and Biomedical Sciences, University of Strathclyde, Glasgow, UK
| | - Somayah S Elsayed
- Department of Molecular Biotechnology, Institute of Biology, Leiden University, Leiden, The Netherlands
| | - Neha Garg
- School of Chemistry and Biochemistry, Center for Microbial Dynamics and Infection, Georgia Institute of Technology, Atlanta, GA, USA
| | - Justin J J van der Hooft
- Bioinformatics Group, Wageningen University, Wageningen, The Netherlands
- Department of Biochemistry, University of Johannesburg, Johannesburg, South Africa
| | - Nathaniel I Martin
- Biological Chemistry Group, Institute of Biology, Leiden University, Leiden, The Netherlands
| | - David Meijer
- Bioinformatics Group, Wageningen University, Wageningen, The Netherlands
| | - Barbara R Terlouw
- Bioinformatics Group, Wageningen University, Wageningen, The Netherlands
| | - Friederike Biermann
- Bioinformatics Group, Wageningen University, Wageningen, The Netherlands
- Institute of Molecular Bio Science, Goethe-University Frankfurt, Frankfurt am Main, Germany
- LOEWE Center for Translational Biodiversity Genomics (TBG), Frankfurt am Main, Germany
| | - Kai Blin
- The Novo Nordisk Foundation Center for Biosustainability, Technical University of Denmark, Kongens Lyngby, Denmark
| | | | - Marina Gorostiola González
- Drug Discovery and Safety, Leiden Academic Centre for Drug Research, Leiden, The Netherlands
- ONCODE institute, Leiden, The Netherlands
| | - Eric J N Helfrich
- Institute of Molecular Bio Science, Goethe-University Frankfurt, Frankfurt am Main, Germany
- LOEWE Center for Translational Biodiversity Genomics (TBG), Frankfurt am Main, Germany
| | - Florian Huber
- Center for Digitalization and Digitality, Hochschule Düsseldorf, Düsseldorf, Germany
| | - Stefan Leopold-Messer
- Institut für Mikrobiologie, Eidgenössische Technische Hochschule (ETH) Zürich, Zürich, Switzerland
| | - Kohulan Rajan
- Institute for Inorganic and Analytical Chemistry, Friedrich-Schiller-University Jena, Jena, Germany
| | - Tristan de Rond
- School of Chemical Sciences, University of Auckland, Auckland, New Zealand
| | - Jeffrey A van Santen
- Department of Chemistry, Simon Fraser University, Burnaby, British Columbia, Canada
| | - Maria Sorokina
- Institute for Inorganic and Analytical Chemistry, Friedrich-Schiller University, Jena, Germany
- Pharmaceuticals R&D, Bayer AG, Berlin, Germany
| | - Marcy J Balunas
- Department of Microbiology and Immunology, University of Michigan, Ann Arbor, MI, USA
- Department of Medicinal Chemistry, University of Michigan, Ann Arbor, MI, USA
| | - Mehdi A Beniddir
- Équipe "Chimie des Substances Naturelles", Université Paris-Saclay, CNRS, BioCIS, Orsay, France
| | - Doris A van Bergeijk
- Department of Molecular Biotechnology, Institute of Biology, Leiden University, Leiden, The Netherlands
| | - Laura M Carroll
- Structural and Computational Biology Unit, EMBL, Heidelberg, Germany
| | - Chase M Clark
- Division of Pharmaceutical Sciences, School of Pharmacy, University of Wisconsin-Madison, Madison, WI, USA
| | | | | | - Chao Du
- Department of Molecular Biotechnology, Institute of Biology, Leiden University, Leiden, The Netherlands
| | | | - Francesca Grisoni
- Institute for Complex Molecular Systems, Department of Biomedical Engineering, Eindhoven University of Technology, Eindhoven, The Netherlands
- Centre for Living Technologies, Alliance TU/e, WUR, UU, UMC Utrecht, Utrecht, The Netherlands
| | | | - Willem Jespers
- Drug Discovery and Safety, Leiden Academic Centre for Drug Research, Leiden, The Netherlands
| | - Olga V Kalinina
- Helmholtz Institute for Pharmaceutical Research Saarland (HIPS), Helmholtz Centre for Infection Research (HZI), Saarbrücken, Germany
- Drug Bioinformatics, Medical Faculty, Saarland University, Homburg, Germany
- Center for Bioinformatics, Saarland University, Saarbrücken, Germany
| | | | - Hyunwoo Kim
- College of Pharmacy and Integrated Research Institute for Drug Development, Dongguk University Seoul, Goyang-si, Republic of Korea
| | - Tiago F Leao
- Center for Nuclear Energy in Agriculture, University of São Paulo, Piracicaba, Brazil
| | - Joleen Masschelein
- Center for Microbiology, VIB-KU Leuven, Heverlee, Belgium
- Department of Biology, KU Leuven, Heverlee, Belgium
| | - Evan R Rees
- Division of Pharmaceutical Sciences, School of Pharmacy, University of Wisconsin-Madison, Madison, WI, USA
| | - Raphael Reher
- Institute of Pharmaceutical Biology and Biotechnology, University of Marburg, Marburg, Germany
- Institute of Pharmacy, Martin-Luther-University Halle-Wittenberg, Halle (Saale), Germany
| | - Daniel Reker
- Department of Biomedical Engineering, Duke University, Durham, NC, USA
- Duke Microbiome Center, Duke University, Durham, NC, USA
| | - Philippe Schwaller
- Laboratory of Artificial Chemical Intelligence, Institut des Sciences et Ingénierie Chimiques, Ecole Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland
| | | | - Michael A Skinnider
- Adapsyn Bioscience, Hamilton, Ontario, Canada
- Michael Smith Laboratories, University of British Columbia, Vancouver, British Columbia, Canada
| | - Allison S Walker
- Department of Chemistry, Vanderbilt University, Nashville, TN, USA
- Department of Biological Sciences, Vanderbilt University, Nashville, TN, USA
| | - Egon L Willighagen
- Department of Bioinformatics - BiGCaT, NUTRIM, Maastricht University, Maastricht, The Netherlands
| | - Barbara Zdrazil
- European Bioinformatics Institute (EMBL-EBI), Wellcome Genome Campus, Cambridgeshire, UK
| | - Nadine Ziemert
- Interfaculty Institute for Microbiology and Infection Medicine Tuebingen (IMIT), Institute for Bioinformatics and Medical Informatics (IBMI), University of Tuebingen, Tuebingen, Germany
| | | | - Pierre Guyomard
- Bonsai team, CRIStAL - Centre de Recherche en Informatique Signal et Automatique de Lille, Université de Lille, Villeneuve d'Ascq Cedex, France
| | - Andrea Volkamer
- Center for Bioinformatics, Saarland University, Saarbrücken, Germany
- In silico Toxicology and Structural Bioinformatics, Institute of Physiology, Charité - Universitätsmedizin Berlin, Berlin, Germany
| | - William H Gerwick
- Scripps Institution of Oceanography, University of California San Diego, La Jolla, CA, USA
| | - Hyun Uk Kim
- Department of Chemical and Biomolecular Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea
| | - Rolf Müller
- Helmholtz Institute for Pharmaceutical Research Saarland (HIPS), Helmholtz Centre for Infection Research (HZI), Saarbrücken, Germany
- Department of Pharmacy, Saarland University, Saarbrücken, Germany
- German Center for infection research (DZIF), Braunschweig, Germany
- Helmholtz International Lab for Anti-Infectives, Saarbrücken, Germany
| | - Gilles P van Wezel
- Department of Molecular Biotechnology, Institute of Biology, Leiden University, Leiden, The Netherlands
- Netherlands Institute of Ecology, NIOO-KNAW, Wageningen, The Netherlands
| | - Gerard J P van Westen
- Drug Discovery and Safety, Leiden Academic Centre for Drug Research, Leiden, The Netherlands.
| | - Anna K H Hirsch
- Helmholtz Institute for Pharmaceutical Research Saarland (HIPS), Helmholtz Centre for Infection Research (HZI), Saarbrücken, Germany.
- Department of Pharmacy, Saarland University, Saarbrücken, Germany.
- German Center for infection research (DZIF), Braunschweig, Germany.
- Helmholtz International Lab for Anti-Infectives, Saarbrücken, Germany.
| | - Roger G Linington
- Department of Chemistry, Simon Fraser University, Burnaby, British Columbia, Canada.
| | - Serina L Robinson
- Department of Environmental Microbiology, Eawag: Swiss Federal Institute for Aquatic Science and Technology, Dübendorf, Switzerland.
| | - Marnix H Medema
- Bioinformatics Group, Wageningen University, Wageningen, The Netherlands.
- Institute of Biology, Leiden University, Leiden, The Netherlands.
| |
Collapse
|
3
|
Seal S, Yang H, Trapotsi MA, Singh S, Carreras-Puigvert J, Spjuth O, Bender A. Merging bioactivity predictions from cell morphology and chemical fingerprint models using similarity to training data. J Cheminform 2023; 15:56. [PMID: 37268960 DOI: 10.1186/s13321-023-00723-x] [Citation(s) in RCA: 9] [Impact Index Per Article: 9.0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/05/2023] [Accepted: 04/20/2023] [Indexed: 06/04/2023] Open
Abstract
The applicability domain of machine learning models trained on structural fingerprints for the prediction of biological endpoints is often limited by the lack of diversity of chemical space of the training data. In this work, we developed similarity-based merger models which combined the outputs of individual models trained on cell morphology (based on Cell Painting) and chemical structure (based on chemical fingerprints) and the structural and morphological similarities of the compounds in the test dataset to compounds in the training dataset. We applied these similarity-based merger models using logistic regression models on the predictions and similarities as features and predicted assay hit calls of 177 assays from ChEMBL, PubChem and the Broad Institute (where the required Cell Painting annotations were available). We found that the similarity-based merger models outperformed other models with an additional 20% assays (79 out of 177 assays) with an AUC > 0.70 compared with 65 out of 177 assays using structural models and 50 out of 177 assays using Cell Painting models. Our results demonstrated that similarity-based merger models combining structure and cell morphology models can more accurately predict a wide range of biological assay outcomes and further expanded the applicability domain by better extrapolating to new structural and morphology spaces.
Collapse
Affiliation(s)
- Srijit Seal
- Yusuf Hamied Department of Chemistry, University of Cambridge, Cambridge, UK
| | - Hongbin Yang
- Yusuf Hamied Department of Chemistry, University of Cambridge, Cambridge, UK
| | - Maria-Anna Trapotsi
- Yusuf Hamied Department of Chemistry, University of Cambridge, Cambridge, UK
| | - Satvik Singh
- Department of Applied Mathematics and Theoretical Physics (DAMTP), University of Cambridge, Cambridge, UK
| | - Jordi Carreras-Puigvert
- Department of Pharmaceutical Biosciences and Science for Life Laboratory, Uppsala University, Uppsala, Sweden
| | - Ola Spjuth
- Department of Pharmaceutical Biosciences and Science for Life Laboratory, Uppsala University, Uppsala, Sweden.
| | - Andreas Bender
- Yusuf Hamied Department of Chemistry, University of Cambridge, Cambridge, UK.
| |
Collapse
|
4
|
Walter M, Allen LN, de la Vega de León A, Webb SJ, Gillet VJ. Analysis of the benefits of imputation models over traditional QSAR models for toxicity prediction. J Cheminform 2022; 14:32. [PMID: 35672779 PMCID: PMC9172131 DOI: 10.1186/s13321-022-00611-w] [Citation(s) in RCA: 8] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/01/2022] [Accepted: 05/12/2022] [Indexed: 11/21/2022] Open
Abstract
Recently, imputation techniques have been adapted to predict activity values among sparse bioactivity matrices, showing improvements in predictive performance over traditional QSAR models. These models are able to use experimental activity values for auxiliary assays when predicting the activity of a test compound on a specific assay. In this study, we tested three different multi-task imputation techniques on three classification-based toxicity datasets: two of small scale (12 assays each) and one large scale with 417 assays. Moreover, we analyzed in detail the improvements shown by the imputation models. We found that test compounds that were dissimilar to training compounds, as well as test compounds with a large number of experimental values for other assays, showed the largest improvements. We also investigated the impact of sparsity on the improvements seen as well as the relatedness of the assays being considered. Our results show that even a small amount of additional information can provide imputation methods with a strong boost in predictive performance over traditional single task and multi-task predictive models.
Collapse
|
5
|
Morger A, Garcia de Lomana M, Norinder U, Svensson F, Kirchmair J, Mathea M, Volkamer A. Studying and mitigating the effects of data drifts on ML model performance at the example of chemical toxicity data. Sci Rep 2022; 12:7244. [PMID: 35508546 PMCID: PMC9068909 DOI: 10.1038/s41598-022-09309-3] [Citation(s) in RCA: 4] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/28/2021] [Accepted: 03/17/2022] [Indexed: 11/09/2022] Open
Abstract
Machine learning models are widely applied to predict molecular properties or the biological activity of small molecules on a specific protein. Models can be integrated in a conformal prediction (CP) framework which adds a calibration step to estimate the confidence of the predictions. CP models present the advantage of ensuring a predefined error rate under the assumption that test and calibration set are exchangeable. In cases where the test data have drifted away from the descriptor space of the training data, or where assay setups have changed, this assumption might not be fulfilled and the models are not guaranteed to be valid. In this study, the performance of internally valid CP models when applied to either newer time-split data or to external data was evaluated. In detail, temporal data drifts were analysed based on twelve datasets from the ChEMBL database. In addition, discrepancies between models trained on publicly-available data and applied to proprietary data for the liver toxicity and MNT in vivo endpoints were investigated. In most cases, a drastic decrease in the validity of the models was observed when applied to the time-split or external (holdout) test sets. To overcome the decrease in model validity, a strategy for updating the calibration set with data more similar to the holdout set was investigated. Updating the calibration set generally improved the validity, restoring it completely to its expected value in many cases. The restored validity is the first requisite for applying the CP models with confidence. However, the increased validity comes at the cost of a decrease in model efficiency, as more predictions are identified as inconclusive. This study presents a strategy to recalibrate CP models to mitigate the effects of data drifts. Updating the calibration sets without having to retrain the model has proven to be a useful approach to restore the validity of most models.
Collapse
Affiliation(s)
- Andrea Morger
- In Silico Toxicology and Structural Bioinformatics, Institute of Physiology, Charité Universitätsmedizin Berlin, Berlin, 10117, Germany
| | - Marina Garcia de Lomana
- BASF SE, 67056, Ludwigshafen, Germany
- Division of Pharmaceutical Chemistry, Department of Pharmaceutical Sciences, University of Vienna, Vienna, 1090, Austria
| | - Ulf Norinder
- Department of Pharmaceutical Biosciences, Uppsala University, Uppsala, 751 24, Sweden
- Dept Computer and Systems Sciences, Stockholm University, Kista, 164 07, Sweden
- MTM Research Centre, School of Science and Technology, 701 82, Örebro, Sweden
| | - Fredrik Svensson
- Alzheimer's Research UK UCL Drug Discovery Institute, London, WC1E 6BT, UK
| | - Johannes Kirchmair
- Division of Pharmaceutical Chemistry, Department of Pharmaceutical Sciences, University of Vienna, Vienna, 1090, Austria
| | | | - Andrea Volkamer
- In Silico Toxicology and Structural Bioinformatics, Institute of Physiology, Charité Universitätsmedizin Berlin, Berlin, 10117, Germany.
| |
Collapse
|
6
|
Oguike OE, Ugwuishiwu CH, Asogwa CN, Nnadi CO, Obonga WO, Attama AA. Systematic review on the application of machine learning to quantitative structure-activity relationship modeling against Plasmodium falciparum. Mol Divers 2022; 26:3447-3462. [PMID: 35064444 PMCID: PMC8782692 DOI: 10.1007/s11030-022-10380-1] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/05/2021] [Accepted: 01/07/2022] [Indexed: 11/29/2022]
Abstract
Malaria accounts for over two million deaths globally. To flatten this curve, there is a need to develop new and high potent drugs against Plasmodium falciparum. Some major challenges include the dearth of suitable animal models for anti-P. falciparum assays, resistance to first-line drugs, lack of vaccines and the complex life cycle of Plasmodium. Gladly, newer approaches to antimalarial drug discovery have emerged due to the release of large datasets by pharmaceutical companies. This review provides insights into these new approaches to drug discovery covering different machine learning tools, which enhance the development of new compounds. It provides a systematic review on the use and prospects of machine learning in predicting, classifying and clustering IC50 values of bioactive compounds against P. falciparum. The authors identified many machine learning tools yet to be applied for this purpose. However, Random Forest and Support Vector Machines have been extensively applied though on a limited dataset of compounds.
Collapse
Affiliation(s)
- Osondu Everestus Oguike
- Machine Learning Research Group, University of Nigeria, Nsukka, 410001, Enugu State, Nigeria.,Department of Computer Science, Faculty of Physical Sciences, University of Nigeria, Nsukka, 410001, Enugu State, Nigeria
| | - Chikodili Helen Ugwuishiwu
- Machine Learning Research Group, University of Nigeria, Nsukka, 410001, Enugu State, Nigeria.,Department of Computer Science, Faculty of Physical Sciences, University of Nigeria, Nsukka, 410001, Enugu State, Nigeria
| | - Caroline Ngozi Asogwa
- Machine Learning Research Group, University of Nigeria, Nsukka, 410001, Enugu State, Nigeria.,Department of Computer Science, Faculty of Physical Sciences, University of Nigeria, Nsukka, 410001, Enugu State, Nigeria
| | - Charles Okeke Nnadi
- Machine Learning Research Group, University of Nigeria, Nsukka, 410001, Enugu State, Nigeria. .,Deprtment of Pharmaceutical and Medicinal Chemistry, Faculty of Pharmaceutical Sciences, University of Nigeria, Nsukka, 410001, Enugu State, Nigeria.
| | - Wilfred Ofem Obonga
- Machine Learning Research Group, University of Nigeria, Nsukka, 410001, Enugu State, Nigeria.,Deprtment of Pharmaceutical and Medicinal Chemistry, Faculty of Pharmaceutical Sciences, University of Nigeria, Nsukka, 410001, Enugu State, Nigeria
| | - Anthony Amaechi Attama
- Machine Learning Research Group, University of Nigeria, Nsukka, 410001, Enugu State, Nigeria.,Department of Pharmaceutics, Faculty of Pharmaceutical Sciences, University of Nigeria, Nsukka, 410001, Enugu State, Nigeria
| |
Collapse
|
7
|
Development and implementation of an enterprise-wide predictive model for early absorption, distribution, metabolism and excretion properties. Future Med Chem 2021; 13:1639-1654. [PMID: 34528444 DOI: 10.4155/fmc-2021-0138] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/31/2022] Open
Abstract
Background: Accurate prediction of absorption, distribution, metabolism and excretion (ADME) properties can facilitate the identification of promising drug candidates. Methodology & Results: The authors present the Janssen generic Target Product Profile (gTPP) model, which predicts 18 early ADME properties, employs a graph convolutional neural network algorithm and was trained on between 1000-10,000 internal data points per predicted parameter. gTPP demonstrated stronger predictive power than pretrained commercial ADME models and automatic model builders. Through a novel logging method, the authors report gTPP usage for more than 200 Janssen drug discovery scientists. Conclusion: The investigators successfully enabled the rapid and systematic implementation of predictive ML tools across a drug discovery pipeline in all therapeutic areas. This experience provides useful guidance for other large-scale AI/ML deployment efforts.
Collapse
|
8
|
Mervin LH, Trapotsi MA, Afzal AM, Barrett IP, Bender A, Engkvist O. Probabilistic Random Forest improves bioactivity predictions close to the classification threshold by taking into account experimental uncertainty. J Cheminform 2021; 13:62. [PMID: 34412708 PMCID: PMC8375213 DOI: 10.1186/s13321-021-00539-7] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/07/2021] [Accepted: 07/30/2021] [Indexed: 11/24/2022] Open
Abstract
Measurements of protein–ligand interactions have reproducibility limits due to experimental errors. Any model based on such assays will consequentially have such unavoidable errors influencing their performance which should ideally be factored into modelling and output predictions, such as the actual standard deviation of experimental measurements (σ) or the associated comparability of activity values between the aggregated heterogenous activity units (i.e., Ki versus IC50 values) during dataset assimilation. However, experimental errors are usually a neglected aspect of model generation. In order to improve upon the current state-of-the-art, we herein present a novel approach toward predicting protein–ligand interactions using a Probabilistic Random Forest (PRF) classifier. The PRF algorithm was applied toward in silico protein target prediction across ~ 550 tasks from ChEMBL and PubChem. Predictions were evaluated by taking into account various scenarios of experimental standard deviations in both training and test sets and performance was assessed using fivefold stratified shuffled splits for validation. The largest benefit in incorporating the experimental deviation in PRF was observed for data points close to the binary threshold boundary, when such information was not considered in any way in the original RF algorithm. For example, in cases when σ ranged between 0.4–0.6 log units and when ideal probability estimates between 0.4–0.6, the PRF outperformed RF with a median absolute error margin of ~ 17%. In comparison, the baseline RF outperformed PRF for cases with high confidence to belong to the active class (far from the binary decision threshold), although the RF models gave errors smaller than the experimental uncertainty, which could indicate that they were overtrained and/or over-confident. Finally, the PRF models trained with putative inactives decreased the performance compared to PRF models without putative inactives and this could be because putative inactives were not assigned an experimental pXC50 value, and therefore they were considered inactives with a low uncertainty (which in practice might not be true). In conclusion, PRF can be useful for target prediction models in particular for data where class boundaries overlap with the measurement uncertainty, and where a substantial part of the training data is located close to the classification threshold.
Collapse
Affiliation(s)
- Lewis H Mervin
- Molecular AI, Discovery Sciences, R&D, AstraZeneca, Cambridge, UK.
| | - Maria-Anna Trapotsi
- Department of Chemistry, Centre for Molecular Informatics, University of Cambridge, Lensfield Road, Cambridge, CB2 1EW, UK
| | - Avid M Afzal
- Data Sciences & Quantitative Biology, Discovery Sciences, R&D, AstraZeneca, Cambridge, UK
| | - Ian P Barrett
- Data Sciences & Quantitative Biology, Discovery Sciences, R&D, AstraZeneca, Cambridge, UK
| | - Andreas Bender
- Department of Chemistry, Centre for Molecular Informatics, University of Cambridge, Lensfield Road, Cambridge, CB2 1EW, UK
| | - Ola Engkvist
- Molecular AI, Discovery Sciences, R&D, AstraZeneca, Gothenburg, Sweden.,Department of Computer Science and Engineering, Chalmers University of Technology, Gothenburg, Sweden
| |
Collapse
|
9
|
Wilm A, Garcia de Lomana M, Stork C, Mathai N, Hirte S, Norinder U, Kühnl J, Kirchmair J. Predicting the Skin Sensitization Potential of Small Molecules with Machine Learning Models Trained on Biologically Meaningful Descriptors. Pharmaceuticals (Basel) 2021; 14:ph14080790. [PMID: 34451887 PMCID: PMC8402010 DOI: 10.3390/ph14080790] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/14/2021] [Revised: 08/03/2021] [Accepted: 08/06/2021] [Indexed: 02/06/2023] Open
Abstract
In recent years, a number of machine learning models for the prediction of the skin sensitization potential of small organic molecules have been reported and become available. These models generally perform well within their applicability domains but, as a result of the use of molecular fingerprints and other non-intuitive descriptors, the interpretability of the existing models is limited. The aim of this work is to develop a strategy to replace the non-intuitive features by predicted outcomes of bioassays. We show that such replacement is indeed possible and that as few as ten interpretable, predicted bioactivities are sufficient to reach competitive performance. On a holdout data set of 257 compounds, the best model (“Skin Doctor CP:Bio”) obtained an efficiency of 0.82 and an MCC of 0.52 (at the significance level of 0.20). Skin Doctor CP:Bio is available free of charge for academic research. The modeling strategies explored in this work are easily transferable and could be adopted for the development of more interpretable machine learning models for the prediction of the bioactivity and toxicity of small organic compounds.
Collapse
Affiliation(s)
- Anke Wilm
- Center for Bioinformatics (ZBH), Department of Informatics, Universität Hamburg, 20146 Hamburg, Germany; (A.W.); (C.S.)
- HITeC e.V., 22527 Hamburg, Germany
| | - Marina Garcia de Lomana
- Department of Pharmaceutical Sciences, Faculty of Life Sciences, University of Vienna, 1090 Vienna, Austria; (M.G.d.L.); (S.H.)
| | - Conrad Stork
- Center for Bioinformatics (ZBH), Department of Informatics, Universität Hamburg, 20146 Hamburg, Germany; (A.W.); (C.S.)
| | - Neann Mathai
- Computational Biology Unit (CBU), Department of Chemistry, University of Bergen, N-5020 Bergen, Norway;
| | - Steffen Hirte
- Department of Pharmaceutical Sciences, Faculty of Life Sciences, University of Vienna, 1090 Vienna, Austria; (M.G.d.L.); (S.H.)
| | - Ulf Norinder
- MTM Research Centre, School of Science and Technology, Örebro University, SE-70182 Örebro, Sweden;
- Department of Computer and Systems Sciences, Stockholm University, SE-16407 Kista, Sweden
- Department of Pharmaceutical Biosciences, Uppsala University, SE-75124 Uppsala, Sweden
| | - Jochen Kühnl
- Front End Innovation, Beiersdorf AG, 22529 Hamburg, Germany;
| | - Johannes Kirchmair
- Center for Bioinformatics (ZBH), Department of Informatics, Universität Hamburg, 20146 Hamburg, Germany; (A.W.); (C.S.)
- Department of Pharmaceutical Sciences, Faculty of Life Sciences, University of Vienna, 1090 Vienna, Austria; (M.G.d.L.); (S.H.)
- Correspondence: ; Tel.: +43-1-4277-55104
| |
Collapse
|
10
|
Garcia de Lomana M, Morger A, Norinder U, Buesen R, Landsiedel R, Volkamer A, Kirchmair J, Mathea M. ChemBioSim: Enhancing Conformal Prediction of In Vivo Toxicity by Use of Predicted Bioactivities. J Chem Inf Model 2021; 61:3255-3272. [PMID: 34153183 PMCID: PMC8317154 DOI: 10.1021/acs.jcim.1c00451] [Citation(s) in RCA: 6] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/20/2021] [Indexed: 02/07/2023]
Abstract
Computational methods such as machine learning approaches have a strong track record of success in predicting the outcomes of in vitro assays. In contrast, their ability to predict in vivo endpoints is more limited due to the high number of parameters and processes that may influence the outcome. Recent studies have shown that the combination of chemical and biological data can yield better models for in vivo endpoints. The ChemBioSim approach presented in this work aims to enhance the performance of conformal prediction models for in vivo endpoints by combining chemical information with (predicted) bioactivity assay outcomes. Three in vivo toxicological endpoints, capturing genotoxic (MNT), hepatic (DILI), and cardiological (DICC) issues, were selected for this study due to their high relevance for the registration and authorization of new compounds. Since the sparsity of available biological assay data is challenging for predictive modeling, predicted bioactivity descriptors were introduced instead. Thus, a machine learning model for each of the 373 collected biological assays was trained and applied on the compounds of the in vivo toxicity data sets. Besides the chemical descriptors (molecular fingerprints and physicochemical properties), these predicted bioactivities served as descriptors for the models of the three in vivo endpoints. For this study, a workflow based on a conformal prediction framework (a method for confidence estimation) built on random forest models was developed. Furthermore, the most relevant chemical and bioactivity descriptors for each in vivo endpoint were preselected with lasso models. The incorporation of bioactivity descriptors increased the mean F1 scores of the MNT model from 0.61 to 0.70 and for the DICC model from 0.72 to 0.82 while the mean efficiencies increased by roughly 0.10 for both endpoints. In contrast, for the DILI endpoint, no significant improvement in model performance was observed. Besides pure performance improvements, an analysis of the most important bioactivity features allowed detection of novel and less intuitive relationships between the predicted biological assay outcomes used as descriptors and the in vivo endpoints. This study presents how the prediction of in vivo toxicity endpoints can be improved by the incorporation of biological information-which is not necessarily captured by chemical descriptors-in an automated workflow without the need for adding experimental workload for the generation of bioactivity descriptors as predicted outcomes of bioactivity assays were utilized. All bioactivity CP models for deriving the predicted bioactivities, as well as the in vivo toxicity CP models, can be freely downloaded from https://doi.org/10.5281/zenodo.4761225.
Collapse
Affiliation(s)
- Marina Garcia de Lomana
- BASF
SE, Ludwigshafen am Rhein 67063, Germany
- Department
of Pharmaceutical Sciences, Faculty of Life Sciences, University of Vienna, Vienna 1090, Austria
| | - Andrea Morger
- In Silico
Toxicology and Structural Bioinformatics, Institute of Physiology, Charité Universitätsmedizin Berlin, Charitéplatz
1, Berlin 10117, Germany
| | - Ulf Norinder
- MTM
Research Centre, School of Science and Technology, Örebro University, Örebro SE-70182, Sweden
| | | | | | - Andrea Volkamer
- In Silico
Toxicology and Structural Bioinformatics, Institute of Physiology, Charité Universitätsmedizin Berlin, Charitéplatz
1, Berlin 10117, Germany
| | - Johannes Kirchmair
- Department
of Pharmaceutical Sciences, Faculty of Life Sciences, University of Vienna, Vienna 1090, Austria
| | | |
Collapse
|