Die Pharmazie is published by IMR Press from Volume 81 Issue 1 (2026). Previous articles were published by another publisher under the CC-BY licence, and they are hosted by IMR Press on imrpress.com as a courtesy and upon agreement.
Prediction of human intestinal absorption using an artificial neural network
X. C. Fu 1, C. X. Chen 2, G. P. Wang 3, Vorname Nachname 4, Vorname Nachname 5
Affiliations
Article Info
1 Department of Pharmacy, Zhejiang University City College, Hangzhou, 310015, P.R. China, Email: Fuxc@zucc.edu.cn
2 First Affiliated Hospital of Zhejiang University, Hangzhou, P.R. China
3 Department of Chemistry, Zhejiang University, Hangzhou, P.R. China
4 College of Pharmaceutical Sciences, Zhejiang University, Hangzhou, P.R. China
5 Ningbo Institute of Technology, Zhejiang University, Ningbo, P.R. China
Abstract
An artificial neural network model is developed to predict percent human intestinal absorption (%FA) of compounds from their molecular structural parameters. These parameters are the polar molecular surface area (PSA), the fraction of polar molecular surface area (FPSA, polar molecular surface area/molecular surface area), the sum of the net atomic charges of oxygen atoms (QO), the sum of the net atomic charges of nitrogen atoms with net negative atomic charges (QN), the sum of the net atomic charges of hydrogen atoms attached to oxygen or nitrogen atoms (QH), and the number of carboxyls (nCOOH). For a training set of 85 compounds and a test set of 10 compounds, root mean squared errors (RMSE) between experimental %FA values and calculated/predicted %FA values are 8.86% and 14.1%, respectively.