Industry-Scale Orchestrated Federated Learning for Drug Discovery

Authors

  • Martijn Oldenhof KU Leuven, ESAT-STADIUS
  • Gergely Ács BME-HIT, CrySyS Lab
  • Balázs Pejó BME-HIT, CrySyS Lab
  • Ansgar Schuffenhauer Novartis Institutes for BioMedical Research
  • Nicholas Holway Novartis Institutes for BioMedical Research
  • Noé Sturm Novartis Institutes for BioMedical Research
  • Arne Dieckmann Bayer AG
  • Oliver Fortmeier Bayer AG
  • Eric Boniface Substra Foundation - Labelia Labs
  • Clément Mayer Substra Foundation - Labelia Labs
  • Arnaud Gohier Institut de recherches Servier
  • Peter Schmidtke Discngine
  • Ritsuya Niwayama Institut de recherches Servier
  • Dieter Kopecky Boehringer Ingelheim RCV GmbH & Co KG
  • Lewis Mervin Molecular AI, Discovery Sciences, R&D, AstraZeneca
  • Prakash Chandra Rathi R&D IT, AstraZeneca
  • Lukas Friedrich Merck KGaA, Global Research & Development
  • András Formanek KU Leuven, ESAT-STADIUS BME-MIT
  • Peter Antal BME-MIT
  • Jordon Rahaman Pillar Biosciences, Inc.
  • Adam Zalewski Amgen Research (Munich) GmbH
  • Wouter Heyndrickx Janssen Pharmaceutica NV
  • Ezron Oluoch Kubermatic
  • Manuel Stößel Kubermatic
  • Michal Vančo Kubermatic
  • David Endico Owkin
  • Fabien Gelus Owkin
  • Thaïs de Boisfossé Owkin
  • Adrien Darbier Owkin
  • Ashley Nicollet Owkin
  • Matthieu Blottière Owkin
  • Maria Telenczuk Owkin
  • Van Tien Nguyen Owkin
  • Thibaud Martinez Owkin
  • Camille Boillet Owkin
  • Kelvin Moutet Owkin
  • Alexandre Picosson Owkin
  • Aurélien Gasser Owkin
  • Inal Djafar Owkin
  • Antoine Simon Owkin
  • Ádám Arany KU Leuven, ESAT-STADIUS
  • Jaak Simm KU Leuven, ESAT-STADIUS
  • Yves Moreau KU Leuven, ESAT-STADIUS
  • Ola Engkvist Molecular AI, Discovery Sciences, R&D, AstraZeneca Department of Computer Science and Engineering, Chalmers University of Technology
  • Hugo Ceulemans Janssen Pharmaceutica NV
  • Camille Marini Owkin
  • Mathieu Galtier Owkin

DOI:

https://doi.org/10.1609/aaai.v37i13.26847

Keywords:

Federated Learning, Drug Discovery, Privacy Preserving, Industry-scale

Abstract

To apply federated learning to drug discovery we developed a novel platform in the context of European Innovative Medicines Initiative (IMI) project MELLODDY (grant n°831472), which was comprised of 10 pharmaceutical companies, academic research labs, large industrial companies and startups. The MELLODDY platform was the first industry-scale platform to enable the creation of a global federated model for drug discovery without sharing the confidential data sets of the individual partners. The federated model was trained on the platform by aggregating the gradients of all contributing partners in a cryptographic, secure way following each training iteration. The platform was deployed on an Amazon Web Services (AWS) multi-account architecture running Kubernetes clusters in private subnets. Organisationally, the roles of the different partners were codified as different rights and permissions on the platform and administrated in a decentralized way. The MELLODDY platform generated new scientific discoveries which are described in a companion paper.
AAAI-23 Proceedings Cover

Downloads

Published

2024-07-15

How to Cite

Oldenhof, M., Ács, G., Pejó, B., Schuffenhauer, A., Holway, N., Sturm, N., Dieckmann, A., Fortmeier, O., Boniface, E., Mayer, C., Gohier, A., Schmidtke, P., Niwayama, R., Kopecky, D., Mervin, L., Rathi, P. C., Friedrich, L., Formanek, A., Antal, P., Rahaman, J., Zalewski, A., Heyndrickx, W., Oluoch, E., Stößel, M., Vančo, M., Endico, D., Gelus, F., de Boisfossé, T., Darbier, A., Nicollet, A., Blottière, M., Telenczuk, M., Nguyen, V. T., Martinez, T., Boillet, C., Moutet, K., Picosson, A., Gasser, A., Djafar, I., Simon, A., Arany, Ádám, Simm, J., Moreau, Y., Engkvist, O., Ceulemans, H., Marini, C., & Galtier, M. (2024). Industry-Scale Orchestrated Federated Learning for Drug Discovery. Proceedings of the AAAI Conference on Artificial Intelligence, 37(13), 15576-15584. https://doi.org/10.1609/aaai.v37i13.26847

Issue

Vol. 37 No. 13: AAAI-23 Special Programs, IAAI-23, EAAI-23, Student Papers and Demonstrations

Section

IAAI Technical Track on deployed Highly Innovative Applications of AI