{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"bioRxiv"}],"indexed":{"date-parts":[[2026,1,15]],"date-time":"2026-01-15T22:30:21Z","timestamp":1768516221819,"version":"3.49.0"},"posted":{"date-parts":[[2024,9,10]]},"group-title":"Biophysics","reference-count":49,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2024,9,10]],"date-time":"2024-09-10T00:00:00Z","timestamp":1725926400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2024,9,10]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:p>Recent advances in machine learning (ML) are reshaping drug discovery. Structure-based ML methods use physically-inspired models to predict binding affinities from protein:ligand complexes. These methods promise to enable the integration of data for many related targets, which addresses issues related to data scarcity for single targets and could enable generalizable predictions for a broad range of targets, including mutants. In this work, we report our experiences in building KinoML, a novel framework for ML in target-based small molecule drug discovery with an emphasis on structure-enabled methods. KinoML focuses currently on kinases as the relative structural conservation of this protein superfamily, particularly in the kinase domain, means it is possible to leverage data from the entire superfamily to make structure-informed predictions about binding affinities, selectivities, and drug resistance. Some key lessons learned in building KinoML include: the importance of reproducible data collection and deposition, the harmonization of molecular data and featurization, and the choice of the right data format to ensure reusability and reproducibility of ML models. As a result, KinoML allows users to easily achieve three tasks: accessing and curating molecular data; featurizing this data with representations suitable for ML applications; and running reproducible ML experiments that require access to ligand, protein, and assay information to predict ligand affinity. Despite KinoML focusing on kinases, this framework can be applied to other proteins. The lessons reported here can help guide the development of platforms for structure-enabled ML in other areas of drug discovery.<\/jats:p>","DOI":"10.1101\/2024.09.10.612176","type":"posted-content","created":{"date-parts":[[2024,9,10]],"date-time":"2024-09-10T15:05:22Z","timestamp":1725980722000},"source":"Crossref","is-referenced-by-count":2,"title":["Lessons learned during the journey of data: from experiment to model for predicting kinase affinity, selectivity, polypharmacology, and resistance"],"prefix":"10.64898","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-2668-7405","authenticated-orcid":false,"given":"Raquel L\u00f3pez-R\u00edos","family":"de Castro","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jaime","family":"Rodr\u00edguez-Guerra","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"David","family":"Schaller","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Talia B.","family":"Kimber","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Corey","family":"Taylor","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jessica B.","family":"White","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Michael","family":"Backenk\u00f6hler","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Alexander","family":"Payne","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ben","family":"Kaminow","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Iv\u00e1n","family":"Pulido","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sukrit","family":"Singh","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Paula Linh","family":"Kramer","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Guillermo","family":"P\u00e9rez-Hern\u00e1ndez","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Andrea","family":"Volkamer","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"John D.","family":"Chodera","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"54368","reference":[{"key":"2024091217352802000_2024.09.10.612176v1.1","doi-asserted-by":"publisher","DOI":"10.1021\/acs.jmedchem.1c00963"},{"key":"2024091217352802000_2024.09.10.612176v1.2","doi-asserted-by":"publisher","DOI":"10.1093\/database\/baab012"},{"key":"2024091217352802000_2024.09.10.612176v1.3","doi-asserted-by":"publisher","DOI":"10.1186\/s12859-016-1433-7"},{"key":"2024091217352802000_2024.09.10.612176v1.4","doi-asserted-by":"publisher","DOI":"10.1016\/j.phrs.2022.106552"},{"key":"2024091217352802000_2024.09.10.612176v1.5","doi-asserted-by":"publisher","DOI":"10.1124\/pr.112.007336"},{"issue":"5","key":"2024091217352802000_2024.09.10.612176v1.6","doi-asserted-by":"crossref","first-page":"509","DOI":"10.4155\/fmc-2015-0003","article-title":"Innovative computer-aided methods for the discovery of new kinase ligands","volume":"8","year":"2016","journal-title":"Future medicinal chemistry"},{"issue":"20","key":"2024091217352802000_2024.09.10.612176v1.7","doi-asserted-by":"crossref","first-page":"4723","DOI":"10.3390\/molecules25204723","article-title":"Merging ligand-based and structure-based methods in drug discovery: An overview of combined virtual screening approaches","volume":"25","year":"2020","journal-title":"Molecules"},{"issue":"17","key":"2024091217352802000_2024.09.10.612176v1.8","doi-asserted-by":"crossref","first-page":"5457","DOI":"10.1021\/acs.jcim.3c00347","article-title":"A Hybrid Structure-Based Machine Learning Approach for Predicting Kinase Inhibition by Small Molecules","volume":"63","year":"2023","journal-title":"Journal of Chemical Information and Modeling"},{"key":"2024091217352802000_2024.09.10.612176v1.9","doi-asserted-by":"publisher","DOI":"10.1021\/acs.jcim.4c00055"},{"key":"2024091217352802000_2024.09.10.612176v1.10","doi-asserted-by":"crossref","unstructured":"Gorantla R , Kubincova A , Wei\u00dfe AY , Mey AS . From Proteins to Ligands: Decoding Deep Learning Methods for Binding Affinity Prediction. Journal of Chemical Information and Modeling. 2023;.","DOI":"10.1101\/2023.08.01.551483"},{"issue":"8","key":"2024091217352802000_2024.09.10.612176v1.11","doi-asserted-by":"crossref","first-page":"3835","DOI":"10.1021\/acs.jcim.1c00653","article-title":"PyRMD: a new fully automated ai-powered ligand-based virtual screening tool","volume":"61","year":"2021","journal-title":"Journal of Chemical Information and Modeling"},{"key":"2024091217352802000_2024.09.10.612176v1.12","doi-asserted-by":"crossref","first-page":"53","DOI":"10.1007\/s10822-006-9096-5","article-title":"Evaluation of machine-learning methods for ligand-based virtual screening","volume":"21","year":"2007","journal-title":"Journal of computer-aided molecular design"},{"key":"2024091217352802000_2024.09.10.612176v1.13","doi-asserted-by":"crossref","unstructured":"Schaller D , Christ CD , Chodera JD , Volkamer A. Benchmarking Cross-Docking Strategies for Structure-Informed Machine Learning in Kinase Drug Discovery. bioRxiv. 2023; p. 2023\u201309.","DOI":"10.1101\/2023.09.11.557138"},{"key":"2024091217352802000_2024.09.10.612176v1.14","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-023-05905-z"},{"issue":"9","key":"2024091217352802000_2024.09.10.612176v1.15","doi-asserted-by":"crossref","first-page":"4435","DOI":"10.3390\/ijms22094435","article-title":"Deep learning in virtual screening: recent applications and developments","volume":"22","year":"2021","journal-title":"International journal of molecular sciences"},{"key":"2024091217352802000_2024.09.10.612176v1.16","doi-asserted-by":"publisher","DOI":"10.26434\/chemrxiv-2023-kz1pb"},{"issue":"12","key":"2024091217352802000_2024.09.10.612176v1.17","doi-asserted-by":"crossref","first-page":"1390","DOI":"10.1038\/s42256-023-00751-0","article-title":"Calibrated geometric deep learning improves kinase\u2013drug binding predictions","volume":"5","year":"2023","journal-title":"Nature Machine Intelligence"},{"key":"2024091217352802000_2024.09.10.612176v1.18","doi-asserted-by":"publisher","DOI":"10.1093\/nar\/gkaa895"},{"key":"2024091217352802000_2024.09.10.612176v1.19","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.1814279116"},{"key":"2024091217352802000_2024.09.10.612176v1.20","doi-asserted-by":"publisher","DOI":"10.1093\/nar\/gkab920"},{"key":"2024091217352802000_2024.09.10.612176v1.21","doi-asserted-by":"publisher","DOI":"10.1093\/nar\/gkv1082"},{"key":"2024091217352802000_2024.09.10.612176v1.22","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0181585"},{"key":"2024091217352802000_2024.09.10.612176v1.23","doi-asserted-by":"crossref","unstructured":"Zdrazil B , Felix E , Hunter F , Manners EJ , Blackshaw J , Corbett S , de Veij M , Ioannidis H , Mendez Lopez D , Mosquera JF , et al. The ChEMBL Database in 2023: a drug discovery platform spanning multiple bioactivity data types and time periods. Nucleic Acids Research. 2023; p. gkad1004.","DOI":"10.1093\/nar\/gkad1004"},{"key":"2024091217352802000_2024.09.10.612176v1.24","unstructured":"Stodden V. Enabling reproducible research: Open licensing for scientific innovation. International Journal of Communications Law and Policy, Forthcoming. 2009; ."},{"issue":"1","key":"2024091217352802000_2024.09.10.612176v1.25","first-page":"1","article-title":"The FAIR Guiding Principles for scientific data management and stewardship","volume":"3","year":"2016","journal-title":"Scientific data"},{"key":"2024091217352802000_2024.09.10.612176v1.26","doi-asserted-by":"publisher","DOI":"10.1126\/science.1075762"},{"key":"2024091217352802000_2024.09.10.612176v1.27","doi-asserted-by":"publisher","DOI":"10.1186\/s12859-016-1452-4"},{"key":"2024091217352802000_2024.09.10.612176v1.28","doi-asserted-by":"publisher","DOI":"10.1093\/nar\/gkab920"},{"key":"2024091217352802000_2024.09.10.612176v1.29","doi-asserted-by":"publisher","DOI":"10.1126\/science.1075762"},{"key":"2024091217352802000_2024.09.10.612176v1.30","doi-asserted-by":"publisher","DOI":"10.1093\/nar\/gkj147"},{"issue":"1","key":"2024091217352802000_2024.09.10.612176v1.31","first-page":"1","article-title":"UniProt tools","volume":"53","year":"2016","journal-title":"Current protocols in bioinformatics"},{"issue":"3","key":"2024091217352802000_2024.09.10.612176v1.32","doi-asserted-by":"crossref","first-page":"e697","DOI":"10.1002\/cpz1.697","article-title":"UniProt Tools: BLAST, Align, Peptide Search, and ID Mapping","volume":"3","year":"2023","journal-title":"Current Protocols"},{"key":"2024091217352802000_2024.09.10.612176v1.33","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0054583"},{"key":"2024091217352802000_2024.09.10.612176v1.34","doi-asserted-by":"publisher","DOI":"10.1021\/jm300131x"},{"key":"2024091217352802000_2024.09.10.612176v1.35","doi-asserted-by":"publisher","DOI":"10.1002\/iub.2060"},{"issue":"20","key":"2024091217352802000_2024.09.10.612176v1.36","doi-asserted-by":"crossref","first-page":"2165","DOI":"10.1021\/acs.biochem.2c00349","article-title":"Cysteine Oxidation in Proteins: Structure, Biophysics, and Simulation","volume":"61","year":"2022","journal-title":"Biochemistry"},{"key":"2024091217352802000_2024.09.10.612176v1.37","doi-asserted-by":"publisher","DOI":"10.1039\/c7mt00204a"},{"key":"2024091217352802000_2024.09.10.612176v1.38","doi-asserted-by":"publisher","DOI":"10.1007\/s00228-018-2580-8"},{"key":"2024091217352802000_2024.09.10.612176v1.39","doi-asserted-by":"publisher","DOI":"10.1186\/s13321-020-00460-5"},{"key":"2024091217352802000_2024.09.10.612176v1.40","doi-asserted-by":"publisher","DOI":"10.1080\/17425255.2021.1951223"},{"key":"2024091217352802000_2024.09.10.612176v1.41","doi-asserted-by":"crossref","unstructured":"Boothroyd S , Behara PK , Madin OC , Hahn DF , Jang H , Gapsys V , Wagner JR , Horton JT , Dotson DL , Thompson MW , et al. Development and Benchmarking of Open Force Field 2.0. 0: The Sage Small Molecule Force Field. Journal of Chemical Theory and Computation. 2023;.","DOI":"10.26434\/chemrxiv-2022-n2z1c-v2"},{"issue":"1-79","key":"2024091217352802000_2024.09.10.612176v1.42","first-page":"4","article-title":"Rdkit documentation","volume":"1","year":"2013","journal-title":"Release"},{"key":"2024091217352802000_2024.09.10.612176v1.43","unstructured":"OpenEye Scientific Software, OEDOCKING 4.2.0.2; 2024. http:\/\/www.eyesopen.com, xinc., Santa Fe, NM."},{"key":"2024091217352802000_2024.09.10.612176v1.44","doi-asserted-by":"crossref","unstructured":"Morris GM , Lim-Wilby M. Molecular docking. Molecular modeling of proteins. 2008; p. 365\u2013382.","DOI":"10.1007\/978-1-59745-177-2_19"},{"issue":"8","key":"2024091217352802000_2024.09.10.612176v1.45","doi-asserted-by":"crossref","first-page":"1771","DOI":"10.1021\/acs.jcim.5b00142","article-title":"POSIT: flexible shape-guided docking for pose prediction","volume":"55","year":"2015","journal-title":"Journal of Chemical Information and Modeling"},{"key":"2024091217352802000_2024.09.10.612176v1.46","doi-asserted-by":"publisher","DOI":"10.1021\/jm051256o"},{"key":"2024091217352802000_2024.09.10.612176v1.47","doi-asserted-by":"publisher","DOI":"10.1126\/science.abo7201"},{"key":"2024091217352802000_2024.09.10.612176v1.48","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-023-05755-9"},{"key":"2024091217352802000_2024.09.10.612176v1.49","doi-asserted-by":"publisher","DOI":"10.7554\/elife.86189"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1101\/2024.09.10.612176","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,15]],"date-time":"2026-01-15T07:01:15Z","timestamp":1768460475000},"score":1,"resource":{"primary":{"URL":"http:\/\/biorxiv.org\/lookup\/doi\/10.1101\/2024.09.10.612176"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,9,10]]},"references-count":49,"URL":"https:\/\/doi.org\/10.1101\/2024.09.10.612176","relation":{},"subject":[],"published":{"date-parts":[[2024,9,10]]},"subtype":"preprint"}}