Premium
Descriptors for amino acids using MolSurf parametrization
Author(s) -
Norinder Ulf,
Svensson Peter
Publication year - 1998
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/(sici)1096-987x(19980115)19:1<51::aid-jcc4>3.0.co;2-y
Subject(s) - parametrization (atmospheric modeling) , ab initio , dipeptide , quantitative structure–activity relationship , set (abstract data type) , computational chemistry , quantum chemical , computer science , quantum , data set , partial least squares regression , algorithm , chemistry , mathematics , data mining , artificial intelligence , amino acid , machine learning , molecule , physics , quantum mechanics , organic chemistry , biochemistry , programming language , radiative transfer
This work describes a new set of amino acid descriptors based on ab initio quantum mechanical calculations and MolSurf technology. These descriptors have been applied to two dipeptide data sets using partial least squares as the statistical engine. Statistically significant models for both data sets have been developed. The results from the derived peptide QSAR models are easy to interpret in terms of the theoretically computed MolSurf parameters of physicochemical nature. © 1998 John Wiley & Sons, Inc. J Comput Chem 19: 51–59, 1998