Jörg Kurt Wegner and Andreas Zell

Prediction of Aqueous Solubility and Partition Coefficient Optimized by a Genetic Algorithm Based Descriptor Selection Method

J. Chem. Inf. Comput. Sci. 2003, 43, pp. 1077-1084

Abstract

The paper describes a fast and flexible descriptor selection method using a genetic algorithm variant (GA-SEC). The relevance of the descriptors will be measured using Shannon entropy (SE) and differential Shannon entropy (DSE), which have very sparse memory requirements and allow the processing of huge data sets. A small quantity of the most important descriptors will be used automatically to build a value prediction model. The most important descriptors are not a linear combination of other descriptors, but transparent, pure descriptors. We used an artificial neural network (ANN) model to predict the aqueous solubility logS and the octanol/water partition coefficient logP. The logS data set was divided into a training set of 1016 compounds and a test set of 253 compounds. A correlation coefficient of 0.93 and an empirical standard deviation of 0.54 were achieved. The logP data set was divided into a training set of 1853 compounds and a test set of 138 compounds. A correlation coefficient of 0.92 and an empirical standard deviation of 0.44 were achieved.

Download

[pdf]

Bibtex 

@Article{wz03,
  author   =     "J. K. Wegner and A. Zell",
  title    =     "{P}rediction of {A}queous {S}olubility and {P}artition {C}oefficient {O}ptimized by a {G}enetic {A}lgorithm {B}ased {D}escriptor {S}election {M}ethod",
  abstract =     "The paper describes a fast and flexible descriptor selection method using a genetic algorithm variant (GA-SEC).
                  The relevance of the descriptors will be measured using Shannon entropy (SE) and differential Shannon
                  entropy (DSE), which have very sparse memory requirements and allow the processing of huge data sets.
                  A small quantity of the most important descriptors will be used automatically to build a value prediction
                  model. The most important descriptors are not a linear combination of other descriptors, but transparent,
                  pure descriptors. We used an artificial neural network (ANN) model to predict the aqueous solubility logS
                  and the octanol/water partition coefficient logP. The logS data set was divided into a training set of 1016
                  compounds and a test set of 253 compounds. A correlation coefficient of 0.93 and an empirical standard
                  deviation of 0.54 were achieved. The logP data set was divided into a training set of 1853 compounds and
                  a test set of 138 compounds. A correlation coefficient of 0.92 and an empirical standard deviation of 0.44
                  were achieved.",
  journal  =     "J. Chem. Inf. Comput. Sci.",
  volume   =     "43",
  year     =     "2003",
  pages    =     "1077-1084",
  url      =     "http://dx.doi.org/10.1021/ci034006u",
  doi      =     "10.1021/ci034006u",
  note     =     "",
  contents =     "feature selection, QSAR, genetic algorithm, clique detection, artificial neural network, solubility, logS, logP",
  topics =       "feature selection, QSAR, genetic algorithm, clique detection, artificial neural network, solubility, logS, logP",
}