CACHE

CRITICAL ASSESSMENT OF COMPUTATIONAL HIT-FINDING EXPERIMENTS

DONATE

  • About
    • WHAT IS CACHE
    • Conferences
  • CACHE News
  • CHALLENGES
    • Challenge #1
      • Announcement
      • Computation methods
      • Preliminary results
      • Final results
    • Challenge #2
      • Announcement
      • Computation methods
      • Preliminary results
      • Final Results
    • Challenge #3
      • Announcement
      • Computation methods
      • Preliminary results
      • Final Results
    • Challenge #4
      • Announcement
      • Computation methods
      • Preliminary results
    • Challenge #5
      • Announcement
      • Computation methods
    • Challenge #6
      • Announcement
    • FAQ
  • PUBLICATIONS
  • CONTACT

Challenge #1

Hit Identification
Method type (check all that applies)
De novo design
Deep learning
Free energy perturbation
High-throughput docking
Machine learning
Physics-based
Hybrid of the above
Hybrid of the above
Description of your approach (min 200 and max 800 words)

First, an ultra-large library of ~4.5B purchasable molecules from Enamine REAL will be parameterized and prepared using our own AI-accelerated quantum-mechanical methods to prepare it for subsequent structure-based automated virtual screening. We will then use our DeepDocking method to rapidly filter the library for top-scoring molecules, utilizing a funnelling approach that we have prospectively validated against SARS-CoV-2 main protease: the initial library will be first screened using DeepDocking-accelerated Glide; the identified molecules will be then screened using Autodock-GPU, and the same process will be then used for ICM docking. In this way, we will identify prospective consensus scoring molecules, providing a set of candidates that is highly enriched with active compounds. Our in-house pipeline will be then used to prioritize compounds for the next refinement step. The best candidates from DeepDocking will be subject to accurate molecular dynamics (MD) and relative free energy (RBFE) simulations, which in turn will be included as training data within an Active Learning (AL) cycle, thus the choice of the next generation of molecules for free energy calculations will be AI-driven in the subsequent loops. For representative molecules from scaffold-clustered diverse set an absolute binding free energy is accessed by MD simulation and used as an initial reference point for relative binding free energies computations with reinforcement learning-directed de novo generative models. Ultimately, top hits will be triaged and prioritised according to predicted binding affinity and combination of physical and ADME/tox properties as predicted by the panel of in-house ML methods.

Method Name
DeepDocking, neural network atomistic force fields, molecular dynamics, thermodynamic integration, free energy methods, de novo generative models
Commercial software packages used

Glide, ICM, OpenEye toolkit, AMBER

Free software packages used

OpenChem, torchani, pytorch, Autodock-GPU

Relevant publications of previous uses by your group of this software/method

Gentile, F., Fernandez, M., Ban, F., Ton, A.-T., Mslati, H., Perez, C. F., Leblanc, E., Yaacoub, J. C., Gleave, J., Stern, A., Wong, B., Jean, F., Strynadka, N., Cherkasov, A. Automated Discovery of Noncovalent Inhibitors of SARS-CoV-2 Main Protease by Consensus Deep Docking of 40 Billion Small Molecules. Chem. Sci. 2021, 12 (48), 15960–15974. https://doi.org/10.1039/d1sc05579h Ton, A.-T., Gentile, F., Hsing, M., Ban, F., Cherkasov, A. Rapid Identification of Potential Inhibitors of SARS- CoV-2 Main Protease by Deep Docking of 1.3 Billion Compounds. Mol. Inform. 2020, 39 (8), 1–18. https://doi.org/10.1002/minf.202000028 Gentile, F., Agrawal, V., Hsing, M., Ton, A. T., Ban, F., Norinder, U., Gleave, M. E., Cherkasov, A. Deep Docking: A Deep Learning Platform for Augmentation of Structure Based Drug Discovery. ACS Cent. Sci. 2020, 6 (6), 939–949. https://doi.org/10.1021/acscentsci.0c00229 M. Korshunova, B. Ginsburg, A. Tropsha, O. Isayev OpenChem: A Deep Learning Toolkit for Computational Chemistry and Drug Design. J. Chem. Inf. Model. 2021, 61, 7-13 https://doi.org/10.1021/acs.jcim.0c00971 X. Gao, F. Ramezanghorbani, O. Isayev, J. S. Smith, A. E. Roitberg. TorchANI: A Free and Open Source PyTorch-Based Deep Learning Implementation of the ANI Neural Network Potentials. Journal of Chemical Information and Modeling 2020, 60, 3408. https://doi.org/10.1021/acs.jcim.0c00451 M. Korshunova, N. Huang, S. Capuzzi, D. S. Radchenko, O. Savych, Y. S. Moroz, C. Wells, T. M. Willson, A. Tropsha, and O. Isayev, “A Bag of Tricks for Automated De Novo Design of Molecules with the Desired Properties: Application to EGFR Inhibitor Discovery,” ChemRxiv, 2021. https://doi.org/10.26434/chemrxiv.14045072.v1 A. Cichonska, B. Ravikumar, R. J Allaway, S. Park, F. Wan, O. Isayev, S. Li, M. Mason, A. Lamb, Z. Tanoli, M. Jeon, S. Kim, M. Popova, S. Capuzzi, J. Zeng, K. Dang, G. Koytiger, J. Kang, C. I. Wells, T. M. Willson, T. I. Oprea, A. Schlessinger, D. H. Drewry, G. Stolovitzky, K. Wennerberg, J. Guinney, T. Aittokallio. Crowdsourced mapping of unexplored target space of kinase inhibitors. Nature Commun. 2021, 12, 3307. https://doi.org/10.1038/s41467-021-23165-1 C. Devereux, J. Smith, K. Davis, K. Barros, R. Zubatyuk, O. Isayev*, A. Roitberg. Extending the Applicability of the ANI Deep Learning Molecular Potential to Sulfur and Halogens. J. Chem. Theory Comput. 2020. 16, 4192. https://doi.org/10.1021/acs.jctc.0c00121

Cache

All rights reserved
v5.47.19.49

Footer first

  • Login
  • Applicant Login
  • Terms of Participation
  • Privacy Policy
  • FAQ
  • Docs
This website is licensed under CC-BY 4.0

Toronto website development by Rebel Trail