Premium
CONCEPTS: New dynamic algorithm for de novo drug suggestion
Author(s) -
Pearlman David A.,
Murcko Mark A.
Publication year - 1993
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/jcc.540141008
Subject(s) - computer science , algorithm
We describe a new method for de novo design of molecules that bind to protein active sites. The method, CONCEPTS (Creation of Novel Compounds by Evaluation of Particles at Target Sites), places a group of atom‐like particles in the site. The particles are free to move within the site to improve binding to the protein. A key innovation of this technique is that covalent connections are made among the particles in a stochastic and dynamically reversible manner. These changes in the topology are either accepted or rejected depending on their ability to improve the total energy of the enzyme–inhibitor complex. The method is applied to two test systems: The FK506 binding protein (FKBP‐12) and HIV‐1 aspartyl protease. In both cases, we are able to predict, de novo , drugs that have striking similarities to known potent inhibitors and that can successfully be used to generate “hits” of the known inhibitors from a data base. © John Wiley & Sons, Inc.