Deaths from drug overdoses have hit record numbers in the US, far exceeding the number of deaths from guns or car accidents. Sixty percent of these deaths are caused by either illegal drugs (e.g., heroin) or legal prescription opioids (e.g., Oxycontin and Percocet). Not only has this drawn the attention of the federal government with specific requests to address the issue by increasing drug tracking, enforcement, and public awareness, but also doctors have changed the way they treat pain, becoming more hesitant to prescribe opioids because of the controversy surrounding their abuse. However, effective treatments for pain management are still lacking for the over 110 million American adults suffering from chronic pain, and a full resolution of the problem will most likely require a molecular level understanding of how these drugs work so as to determine how to fine-tune opioid signaling towards the desired therapeutic pathways and away from those mediating adverse effects. This information is essential to eventually design powerful chemical tools that may be developed into improved therapeutics. Building upon the growing body of evidence suggesting that opioid allosteric modulators and G protein-biased opioid agonists may act as improved painkillers with reduced side effects, the overall goal of the work described in this application is to obtain atomistic information about alternative (allosteric) binding sites and/or ligand-specific (biased) conformations of opioid receptors, including details of drug-receptor binding kinetics, for use as new, exciting avenues to eventually discover improved therapeutics. To this end, we will employ a computation-driven approach for hypothesis generation of unique mechanistic insights into opioid receptor function that will be tested iteratively by independently funded investigators. These studies are expected to significantly impact biomedical research by helping us establish the value of opioid allosteric modulation, biased agonism, and binding kinetics in drug discovery.