Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2020 Jun 30:11:770.
doi: 10.3389/fphar.2020.00770. eCollection 2020.

Accelerating Therapeutics for Opportunities in Medicine: A Paradigm Shift in Drug Discovery

Affiliations

Accelerating Therapeutics for Opportunities in Medicine: A Paradigm Shift in Drug Discovery

Izumi V Hinkson et al. Front Pharmacol. .

Abstract

Conventional drug discovery is long and costly, and suffers from high attrition rates, often leaving patients with limited or expensive treatment options. Recognizing the overwhelming need to accelerate this process and increase success, the ATOM consortium was formed by government, industry, and academic partners in October 2017. ATOM applies a team science and open-source approach to foster a paradigm shift in drug discovery. ATOM is developing and validating a precompetitive, preclinical, small molecule drug discovery platform that simultaneously optimizes pharmacokinetics, toxicity, protein-ligand interactions, systems-level models, molecular design, and novel compound generation. To achieve this, the ATOM Modeling Pipeline (AMPL) has been developed to enable advanced and emerging machine learning (ML) approaches to build models from diverse historical drug discovery data. This modular pipeline has been designed to couple with a generative algorithm that optimizes multiple parameters necessary for drug discovery. ATOM's approach is to consider the full pharmacology and therapeutic window of the drug concurrently, through computationally-driven design, thereby reducing the number of molecules that are selected for experimental validation. Here, we discuss the role of collaborative efforts such as consortia and public-private partnerships in accelerating cross disciplinary innovation and the development of open-source tools for drug discovery.

Keywords: artificial intelligence; data science; drug discovery and development; in silico modeling; machine learning.

PubMed Disclaimer

Figures

Figure 1
Figure 1
The ATOM preclinical drug discovery workflow. ATOM is developing an active learning drug discovery framework that uses a compound library as input to a property prediction pipeline. The pipeline begins with historic data collected on a working compound library to train machine learning-based models for property prediction. Next, multi-level and systems-level models of efficacy, safety, and pharmacokinetics as well as developability are integrated to generate a set of drug design criteria. These parameters are simultaneously optimized for the generation of novel molecules by the generative molecular design framework. The multi-parameter optimization loop, in grey, can be run for numerous cycles. An active learning approach is used to decide whether a molecular simulation or experiment is needed to improve or validate the models. Data that result from these simulations and experiments are then used to re-train the property prediction models. The result of this workflow is a set of optimized drug candidates.

References

    1. Alteri E., Guizzaro L. (2018). Be open about drug failures to speed up research. Nature 563 (7731), 317–319. 10.1038/d41586-018-07352-7 - DOI - PubMed
    1. Altshuler J. S., Balogh E., Barker A. D., Eck S. L., Friend S. H., Ginsburg G. S., et al. (2010). Opening up to precompetitive collaboration. Sci. Transl. Med. 2 (52), 52cm26. 10.1126/scitranslmed.3001515 - DOI - PubMed
    1. Berthold M. R. E. A. (2008). “KNIME: The Konstanz Information Miner,” in Data Analysis, Machine Learning and Applications. Studies in Classification, Data Analysis, and Knowledge Organization. Eds. Preisach C. B. H., Schmidt-Thieme L., Decker R. (Berlin, Heidelberg: Springer; ).
    1. Besnard J., Ruda G. F., Setola V., Abecassis K., Rodriguiz R. M., Huang X.-P., et al. (2012). Automated design of ligands to polypharmacological profiles. Nature 492 (7428), 215–220. 10.1038/nature11691 - DOI - PMC - PubMed
    1. Blaschke T., Olivecrona M., Engkvist O., Bajorath J., Chen H. (2018). Application of Generative Autoencoder in De Novo Molecular Design. Mol. Inf. 37 (1-2), 1700123. 10.1002/minf.201700123 - DOI - PMC - PubMed

LinkOut - more resources