NRF2 drug repurposing using a question-answer artificial intelligence system

Abstract

Drug repurposing represents an innovative approach to reduce the drug development timeline. Text mining using artificial intelligence methods offers great potential in the context of drug repurposing. Here, we present a question-answer artificial intelligence (QAAI) system that is capable of repurposing drug compounds. Our system employs a Google semantic AI universal encoder to compute the sentence embedding of an imposed text question in relation to publications stored in our RedBrain JSON database. Sentences similarity is calculated using a sorting function to identify drug compounds. We demonstrate our systems ability to predict new indications for already existing drugs. Activation of the NRF2 pathway seems critical for enhancing several diseases prognosis. We experimentally validated the prediction for the lipoxygenase inhibitor drug zileuton as a modulator of the NRF2 pathway in vitro, with potential applications to reduce macrophage M1 phenotype and ROS production. This novel computational method provides a new approach to reposition of known drugs in order to treat neurodegenerative diseases. Github for the database and the code can be downloaded from https://gist.github.com/micheledw/5a165b44345d45105d715340b88c756b