COVID Synergy: A Machine Learning Approach Uncovering Potential Treatment Combinations for SARS-CoV-2
Abstract
For more than two years, the COVID-19 pandemic has upended the lives of billions of individuals worldwide leading to disruptions in healthcare, the economy and society at large. As the pandemic enters its third year, the human impact cannot be overstated and the need to develop effective pharmaceuticals remains. Though there currently exits FDA-approved medications for COVID-19, the emergence of novel variants, such as Omicron, highlights the importance of discovering new therapies which will continue to be effective regardless of the pandemic’s progression. Because discovering new medications is a costly and time-intensive endeavor, my approach entails drug repurposing to test medications which are already in use. In this publication, combinations of previously approved drugs are tested for synergy against SARS-CoV-2. The intention of using combinations of drugs is to improve patient outcomes and prevent treatment escape. My approach uses various machine learning models to predict synergy for repurposed drugs which have been previously shown to have activity against SARS-CoV-2. Drug synergy models are made publicly available to researchers hoping to study SARS-CoV-2. In addition to the in silico experiments, top-scoring combinations are experimentally validated in vitro.
Subject Area
Pharmaceutical sciences|Artificial intelligence|Computer science
Recommended Citation
Sánchez Orozco, Jason Edén, "COVID Synergy: A Machine Learning Approach Uncovering Potential Treatment Combinations for SARS-CoV-2" (2022). ETD Collection for University of Texas, El Paso. AAI29209983.
https://scholarworks.utep.edu/dissertations/AAI29209983