Opioids are efficacious analgesic medicines but have associated risk of drug addiction and other severe side effects. New opioids are under development with the view to discover safer analgesic drugs. Compared to traditional experimental approaches, computational modeling is a promising strategy to efficiently prioritize new candidates with low cost. In this study, we developed an automatic data mining and computational modeling workflow that was proven to be applicable for new analgesic opioids screening. To this end, a large opioids dataset consisting of 3,656 compounds was used as the probe to automatically obtain bioassay data from the PubChem portal. 114 PubChem assays were selected to build Quantitative Structure-Activity Relationship (QSAR) models based on the testing results across the probe compounds. A model was selected if it has acceptable model performance and has reasonable predictions distribution after predicting untested probe opioids. In total, 49 models developed for 14 bioassays were selected based on the criteria and were identified to be mainly associated with binding affinities to different opioid receptors. The models for these 14 bioassays were used to fill data gaps in the probe opioids dataset and to predict general drug compounds in the DrugBank dataset. This workflow efficiently extracted bioassays suitable for predictive model development, and the resulted virtual bioprofiles could be used to prioritize untested opioids with desired activity profiles for further experimental testing. This study provides a universal modeling strategy that can take advantage of large public datasets for computer-aided drug design (CADD).

Construction of a virtual opioid bioprofile: A data-driven automatic QSAR modeling study to identify new analgesic opioids

ACS Spring Meeting 2021

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.