Premium
Prospective Prediction of Antitarget Activity by Matched Molecular Pairs Analysis
Author(s) -
Warner Daniel J.,
BridglandTaylor Matthew H.,
Sefton Clare E.,
Wood David J.
Publication year - 2012
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.201200020
Subject(s) - quantitative structure–activity relationship , computer science , aggregate (composite) , data mining , machine learning , nanotechnology , materials science
Matched molecular pairs analysis (MMPA)1,2 is an inverse quantitative structure activity relationship (QSAR) technique that is rapidly gaining popularity in the retrospective analysis of large experimental datasets.3,4 While much of the recent focus has been on the differences in properties between structurally related groups of existing compounds, attempts to extend this methodology to the de‐novo design of novel structures have been limited. To our knowledge the aggregate effect of multiple transformations, all suggesting the same molecular structure, has only ever being considered within a very limited dataset.5 We therefore sought to test this exciting new approach to the design (and absolute property prediction – effectively QSAR‐by‐MMPA) of novel chemical entities based on a larger, more diverse dataset, and couple these designs to MMPA‐based predictions of antitarget activity.