Discovery of chemical compound groups with common structures by a network analysis approach (affinity prediction...

Area of Science:

• Medicinal Chemistry
• Computational Chemistry
• Network Science

Background:

• Drug discovery relies on identifying and optimizing lead compounds.
• Traditional lead optimization can be empirical and time-consuming.
• Understanding structure-activity relationships is crucial for efficient optimization.

Purpose of the Study:

• To develop a network-based method for grouping chemical compounds.
• To improve the efficiency of the drug discovery process by automating compound classification.
• To extract meaningful chemical structure patterns from active inhibitors.

Main Methods:

• Characterizing chemical compounds using secondary dimensional descriptors.
• Applying network inference to determine relationships between compounds.
• Utilizing network clustering to group compounds into distinct clusters.
• Analyzing structural commonalities and differences within and between clusters.

Main Results:

• Successfully applied the method to 279 active inhibitors of factor Xa.
• Generated a network of 266 compounds with 408 edges.
• Divided the network into 10 distinct clusters.
• Identified common chemical structures within clusters and diverse structures between clusters.

Conclusions:

• The network approach effectively groups chemical compounds based on their relationships.
• Extracted structural patterns provide rational guidance for lead optimization.
• This method enhances the efficiency of drug discovery by moving beyond empirical approaches.