Graph mining: procedure, application to drug discovery and recent advances.

Area of Science:

• Computational Chemistry
• Graph Theory
• Cheminformatics

Background:

• Combinatorial chemistry generates vast chemical libraries containing potential drug candidates.
• Molecular structures are often represented as graphs, amenable to computational analysis.
• Graph mining offers powerful techniques for analyzing complex chemical structures.

Purpose of the Study:

• To explain the principles and methodology of frequent subgraph mining.
• To illustrate the application of frequent subgraph mining in analyzing molecular graphs.
• To review recent advancements in graph mining for chemical data.

Main Methods:

• Representing chemical compounds as molecular graphs.
• Applying frequent subgraph mining algorithms to identify recurring structural patterns.
• Analyzing the retrieved subgraphs for their significance in drug discovery.

Main Results:

• Identification of essential structural fragments (frequent subgraphs) within large chemical libraries.
• Demonstration of practical applications of subgraph mining in cheminformatics.
• Overview of current trends and innovations in graph mining techniques.

Conclusions:

• Frequent subgraph mining is a valuable tool for extracting meaningful information from molecular graph data.
• This approach aids in the analysis of large chemical databases for drug discovery.
• Continued advances in graph mining enhance its utility in computational chemistry.