Resource- and time-consuming biological experiments are unavoidable in traditional drug development, which directly drives the evolution of computational tools and algorithms for target prediction and virtual screening. Although structure-based, ligand-based, and data-driven approaches have been developed separately by different groups, a platform integrated the three approaches together is still not available to date. In this study, we integrated them together to build a web server, namely D3CARP, for predicting potential binding proteins or screening hit compounds, with the accuracy ranging from 0.44 to 1.00 for each method. The database for establishing the platform has nearly 1 million active compounds together with 5901 associated targets based on which both the ligand-based and data-driven models were constructed, and has 1970 targets with experimentally determined structures based on which a total of 9352 docking models were developed. For maximizing the potential application of the platform, the diseases information with related protein targets was annotated in detail. Especially, the D3CARP provides feasibility for comprehensive analysis of the prediction results from multiple different methods to overcome the shortcomings of the individual computational approach, which is beneficial for the improvement of the prediction accuracy. The web server is freely accessible at https://www.d3pharma.com/D3CARP/index.php.

D3CARP: a platform integrating structure-based, ligand-based and data-driven approaches for target prediction and virtual screening

Where thousands of chemistry professionals meet to share ideas and advance scientific and technical knowledge. Spring 2023 abstracts, posters, and presentations will launch at the end of March.

ACS Spring 2023

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