Machine Learning to Predict Anti-Malarial Compounds

Santos Navarro, Dr. Manoj T. Duraisingh, and Dr. Aditya S. Paul (Mentor), Department of Immunology and Infectious Diseases, Harvard T.H. Chan School of Public Health, 677 Huntington Ave, Boston, MA 02115

We need to better streamline anti-malarial drug discovery in response to the emerging drug resistance in malaria parasites. We used an Artificial Intelligence (AI) program ChemProps to predict new anti-malarial compounds. The neural network program relies on training, which I carried out using a chemically diverse set of 20,000 established antimalarial compounds previously reported in the literature. A high positive rate (70%) observed with implementation of the program in known antimalarial compounds suggests that with training, ChemProps can capture broad features of antimalarials. In the Broad Institute’s Drug Repurposing Hub catalog of 7,000 compounds, the program predicted ~25% to be anti-malarial. My preliminary findings, obtained with limited computing capacity, suggests that AI-based approaches may prove powerful as a mode of screening compounds for antimalarial inhibitors.

Additional Abstract Information

Presenter: Jose Navarro

Institution: University of Colorado at Boulder

Type: Oral

Subject: Computer Science

Status: Approved

Time and Location

Session: Oral 1
Date/Time: Mon 1:30pm-2:30pm
Session Number: 112
List other presenters in this same room and session