Premium
CS‐MINER: A Tool for Association Mining in Binding‐Database
Author(s) -
ManiVarnosfaderani Ahmad,
Valadkhani Abolfazl,
JalaliHeravi Mehdi
Publication year - 2015
Publication title -
molecular informatics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.481
H-Index - 68
eISSN - 1868-1751
pISSN - 1868-1743
DOI - 10.1002/minf.201400142
Subject(s) - association rule learning , computer science , data mining , database , cheminformatics , computational biology , chemistry , biology , computational chemistry
This paper introduces the algorithms, implementation strategies, features, and applications of CS‐MINER, a tool for visualization and analysis of drug‐like chemical space. The CS‐MINER is the abstract abbreviation for Chemical Space Miner and correlates the medicinal target space and chemical space, in a systematic way. The database in this software consists of a large collection of drug‐like molecules. To prepare this database, a large number of molecules for 110 important biological targets were collected from Binding‐DB. A total of 1497 physicochemical properties were calculated for each molecule. The CS‐MINER uses the discriminant analysis techniques for tracing the collected data and finally separates the molecules based on their therapeutic targets and activities. The developed multivariate classifiers can be used for ligand‐based virtual screening of more than 0.5 million random molecules of PubChem and ZINC databases. In order to validate the models, selected subspaces in CS‐MINER were compared with DrugBank molecules. At the end of the analysis, the software provides an interactive environment for visualization of the selected chemical subspaces in the form of 2‐ and 3‐dimensional plots. In general, CS‐MINER is a tool for comparing the relative position of active biosimilar molecules in chemical space and is freely available at www.csminer.com.