Open AccessBoosting n-octanol/water Partition Coefficients Prediction with An Improved Gene Expression Programming Method
Author(s) -
Cheng Yuan,
Yuzhong Peng,
Chao Deng,
Daoqing Gong,
Ang Cao
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1486/4/042042
Subject(s) - gene expression programming , partition coefficient , regression , octanol , quantitative structure–activity relationship , computer science , artificial intelligence , chemistry , mathematics , biological system , machine learning , statistics , chromatography , biology
n -octanol/water partition coefficient (named logP) reflects the lipid solubility and aqueous solubility of the substance. Accurate and effective prediction of logP has great significance for drug development and monitoring human health, due to logP is related to the dissolution, absorption, distribution and transport of the drug in the human body. This study proposed an improved gene expression programming algorithm based on fuzzy control method with the feature of Morgan fingerprint to improve the logP prediction. Experimental results evaluated in terms of RMSE and MAE show that proposed method outperforms not only multicellular gene expression programming, but also the state-of-the-art methods including Back Propagation neural network, support vector regression, random forest regression and Gaussian process regression.