Anvita Gupta

2015 Global Healthcare Challenge, Grand Prize Winner

Anvita Gupta (An-vee-tah Gooptah), a senior from BASIS Scottsdale High School in Scottsdale, AZ has won the 2015 Global Healthcare Challenge, a competition for high school students who demonstrate an exemplary understanding of health care and medical biotechnology through science research projects. Fourteen students from the U.S., Canada and Germany competed in this year’s Challenge. Anvita received an award of $7,500.

Winners were announced Tuesday, June 16, 2015 at the keynote luncheon at the 2015 BIO International Convention, the largest global event for the biotechnology industry.

Gupta’s project titled, “Novel Therapeutics for Genetic and Infectious Diseases: Drug Discovery Targeting Intrinsically Disordered Proteins” won her the challenge.

Students were evaluated on the quality of their research and display, their responses to questions demonstrating their scientific knowledge, and the potential commercial applications of their research.

Please see below for more information regarding Gupta’s project.


Drug discovery takes ten years and five billion dollars to bring one drug to market. By combining artificial-intelligence with biochemistry, we create a novel framework to find therapeutics for genetic/infectious diseases by targeting disordered proteins, thus reducing time and cost. The framework is applied to find and validate new drugs for Cancer, Tuberculosis, and Ebola.


Intrinsically Disordered Proteins (IDPs) constantly change tertiary structure and play an important role in disease pathways. However, IDPs are difficult to target because their changing structure prevents drugs from binding to them. Here, we develop a novel method to block IDP interactions, by finding drugs that bind to the IDP’s protein partners instead of the IDP. An IDP protein partner is another ordered protein that binds to the IDP. The proposed framework, based on machine learning algorithm, compares drugs to the IDP binding site (MoRF) using four 3D similarity features, and predicts whether the drug will be active in blocking IDP interactions. The framework predicted test set cases with >90% accuracy, and was validated through application to the cancer protein thrombin. The classifier ranked FDA approved thrombin inhibitors first out of 3000 drugs. We applied our framework to identify inhibitors of tuberculosis protein targets. We experimentally validated four drug candidates against Mycobacterium tuberculosis. The framework was applied to discover nine potential antivirals to inhibit Ebola’s glycoprotein. We developed the largest known database linking IDPs with associated diseases and drugs. With big-data analytics, we identified four proteins common to breast, pancreatic, and ovarian cancer, of which three have no approved drugs. 13,000 drugs were repurposed for PTEN, one of these proteins, with our classifier. The framework helps reduce the number of bioassays that must be conducted, and the time and cost of translational drug discovery. This framework can rationally find drugs for any genetic/infectious disease in which IDPs are involved.