Document is current

Article History

Accepted: 9 February 2022

First Online: 17 May 2022

Declarations

:

: Research support from MIT Laboratory for Financial Engineering is gratefully acknowledged. No direct funding was received for this study and no funding bodies had any role in the study design, data collection and analysis, decision to publish, or preparation of this article. The authors were personally salaried by their institutions during the period of writing (though no specific salary was set aside or given for the writing of this manuscript).

: Q.X. reports personal investments in publicly traded pharmaceutical companies. E.A. and A.A. are co-founders of Themis AI. D.R. reports personal investments in technology companies and mutual funds. D.R. is co-founder of Venti Technologies, ThemisAI, and The Routing Company. She is a member of the technology advisory board of British Telecom, Hyundai Motor Company, RobGlobal, Knowledge AI, Ten63, Venti, and RIIID. She is a member of the board of trustees of MBZUAI, a senior visiting fellow at The MITRE Corporation, and a member of Accenture’s Luminary program. She was a member of PCAST and DIB and has given recent talks at GITEX, TransformAI, Stavros Niarchos Foundation, Purdue, Harvard Radcliff, UIUC, University of Pennsylvania, Johns Hopkins University, ETH, EPFL, KTH, and University of Cambridge. D.R.’s research is funded by the USA Air Force, NSF, ONR, DARPA, Toyota Research Institute, IBM, The Boeing Company, Amazon, JPMC, DSTA, DSO, GIST, the Israel Ministry of Defense, AMS, SMART, and a TED Audacious Prize. A.W.L. reports personal investments in private biotech companies, biotech venture capital funds, and mutual funds. A.W.L. is a co-founder and partner of QLS Advisors, a healthcare analytics and consulting company; an advisor to Apricity Health, Aracari Bio, BrightEdge Impact Fund, Enable Medicine, FINRA, Lazard, Quantile Health, SalioGen Therapeutics, the Swiss Finance Institute, Thalēs, and Think Therapeutics; a director of AbCellera, Atomwise, BridgeBio Pharma, Roivant Sciences, and Annual Reviews; and a member of the NIH’s National Center for Advancing Translational Sciences Advisory Council. During the most recent 6-year period, A.W.L. has received speaking/consulting fees, honoraria, or other forms of compensation from: AlphaSimplex Group, Annual Reviews, the Bernstein Fabozzi Jacobs Levy Award, BIS, BridgeBio Pharma, Cambridge Associates, Chicago Mercantile Exchange, Financial Times, Harvard Kennedy School, IMF, JOIM, National Bank of Belgium, New Frontiers Advisors (for 2020 Harry M. Markowitz Prize), Q Group, Research Affiliates, Roivant Sciences, and the Swiss Finance Institute.

: Not applicable.

: Not applicable.

: Not applicable.

: Citeline is a proprietary dataset provided by Informa Pharma Intelligence and can be accessed via commercial license. The probability of success estimates extracted from historical drug development data in Citeline can be accessed via Project ALPHA .

: The code for this work is available from the corresponding author on reasonable request.

: The study was first conceived by D.R. and A.W.L. Data curation and preprocessing were performed by Q.X. The code for training the deep learning models, data visualization, and hyperparameter sensitivity analysis was written by E.A., with help from Q.X. and A.A. The deep learning models were trained by Q.X. on the MIT Engaging cluster. The code for formal analysis was primarily written by Q.X., with crucial inputs from E.A. and A.A. and supervision by D.R. and A.W.L. The first draft of the manuscript was written by Q.X. and E.A. All authors reviewed and commented on previous versions of the manuscript. All authors read and approved the final version of the manuscript.