AtlasCBS: a web server to map and explore chemico-biological space

Area of Science:

• Medicinal Chemistry
• Computational Chemistry
• Drug Discovery

Background:

• Small-molecule drug discovery faces high failure rates, necessitating improved methods for compound selection.
• Ligand Efficiency Indices (LEI) are emerging as powerful metrics for early-stage compound evaluation.
• Effective visualization tools are crucial for interpreting complex biological and chemical data.

Purpose of the Study:

• To introduce the AtlasCBS server, a web tool for visualizing drug discovery data.
• To demonstrate how Ligand Efficiency Indices can be used to represent target-ligand interactions.
• To facilitate the selection and prioritization of promising drug candidates.

Main Methods:

• Development of the AtlasCBS server, which processes target-ligand databases.
• Generation of two-dimensional dynamical representations based on Ligand Efficiency Indices.
• Integration of public databases (BindingDB, PDBBind, ChEMBL) and support for proprietary datasets.

Main Results:

• AtlasCBS provides a unique atlas-like representation of compound data using LEI.
• The server effectively decouples chemical (angular) and biological (radial) properties of ligands.
• Visualizations aid in understanding structure-activity relationships and prioritizing compounds.

Conclusions:

• The AtlasCBS server offers a valuable resource for enhancing early-stage drug discovery.
• LEI-based visualizations represent a significant advancement in drug design strategies.
• This approach holds promise for reducing failure rates in small-molecule drug development.