Premium
Design of screening libraries biased for pharmaceutical discovery
Author(s) -
Koehler Ryan T.,
Villar Hugo O.
Publication year - 2000
Publication title -
journal of computational chemistry
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.907
H-Index - 188
eISSN - 1096-987X
pISSN - 0192-8651
DOI - 10.1002/1096-987x(200010)21:13<1145::aid-jcc3>3.0.co;2-t
Subject(s) - virtual screening , identification (biology) , computer science , set (abstract data type) , drug discovery , process (computing) , domain (mathematical analysis) , chemical library , information retrieval , data mining , chemistry , mathematics , small molecule , biology , programming language , mathematical analysis , botany , biochemistry
Screening of increasingly large chemical libraries has become the method of choice in the search for compounds that could be starting points in the pharmaceutical discovery process. As the process matures, the requirements on the compounds that constitute such libraries are becoming more stringent. We have applied the recently developed LASSOO algorithm to augment the structural diversity of a chemical library. The algorithm is based on the identification of compounds from a virtual library that are most different from those already present in a screening set and to a reference set of undesirable compounds, while being simultaneously most similar to a set of compounds with desirable characteristics. We illustrate the concept using chemical databases in the public domain represented by bitstring structure descriptors. © 2000 John Wiley & Sons, Inc. J Comput Chem 21: 1145–1152, 2000