We are pleased to announce that our two published models developed by Dr. Sehan Lee who is a former team leader in the Chemoinformatics Team are going to be part of PreADMET.

Generalized Solvation Free Energy Density (G-SFED)

– Generalized Solvation Free Energy Density (G-SFED) is a generalized version of the SFED with the purpose of predicting solvation free energies in virtually any solvent. In the model, the solvation free energy of a solute is represented as a linear combination of empirical functions of the solute properties representing the effects of solute on various solute-solvent interactions, and the complementary solvent effects on these interactions were reflected in the linear expansion coefficients with a few solvent properties.

G-SFED provides accurate prediction results for a wide range of sizes and polarities of solute molecules in various solvents as shown by a set of5,753 solvation free energies of diverse solute-solvent combinations as well as octanol-water partition coefficients of small organic compounds and peptides.

Human Nephrotoxicity

– Human nephrotoxicity prediction models provide prediction of three common patterns of drug-induced kidney injury, i.e., tubular necrosis, interstitial nephritis, and tubulo-interstitial nephritis. A Support Vector Machine (SVM) with clinical data on the nephrotoxicity of pharmacological compounds was used to build the binary classification models of nephrotoxin versus non-nephrotoxin with eight fingerprint descriptors.

The extended version will be available in Aug 2017.



G-SFED (https://dx.doi.org/10.1073/pnas.1221940110)

Human Nephrotoxicity (http://pubs.acs.org/doi/abs/10.1021/tx400249t)

We are pleased to introduce a new design for PreADMET web site.

Presently, we only support internal explorer brower! As soon as possible, we will do best to support other blower such as firefox…



In 2004, our PreADME was submitted to the 15th European Symposium on Quantitative Structure – Activity relationships & Molecular Modeling (5th – 11th September 2004) at Istanbul, Terkey.

This conference has been designed to review current practices used to identify new lead compounds and optimize their performance in industrial screens and as potential new products.





Sung Kwang Lee1-2, Gue Su Chang2, In Hye Lee2, Jae Eun Chung2, Kil Yean Sung2, Kyoung Tai No2



1 Computer-Chemie-Centrum, University of Erlangen-Nürnberg, Nägelsbachstrasse 25, 91052, Erlangen, Germany
2 Research Institute of Bioinformatics & Molecular Design(BMD), Sinchon-dong 134, Seodaimoon-gu, 120-749,Seoul, Korea


Around half of all drugs in clinical development fail to make it to the market because of poor ADME and toxicity properties. Consequently, there is increasing interest in the early prediction of ADME properties, with the objective of increasing the success rate of compounds reaching development.

We describe a new pc-based program called PreADME, which had developed rapidly and reliable methods to predict drug-likeness and ADME properties. The program can calculate over 900 molecular descriptors including constitutional, topological, electrostatic, physico-chemical and geometrical descriptors to predict ADME properties. Especially, logP and logS in pure water and buffer solution, boiling point, melting point and vapor pressure could be predicted by 210 new atomic types addition(ATA) method.

Using drug-likeness prediction module, it is possible to use Lipinski’s rule, Lead-like rule that most well known rule relating the chemical structures to drug-likeness. Also, it is possible to provide any drug-like methods such as MDDR-like, CMC-like, WDI-like, or advanced customization option.

In drug absorption (Caco-2 cell, MDCK cell, blood-brain barrier, human intestinal absorption and skin permeability) and plasma protein binding prediction, genetic functional approximation is used to select representative descriptors for prediction, following back-propagation neural network to develop successful nonlinear model.

Furthermore we are working on collecting databases containing ADME and toxicity data such as Vd, Km, t1/2, CLunbound, urinary excretion and mutagenicity data in order to train the network for physiological based pharmacokinetics (PBPK) model and toxicity prediction. And some physicochemical properties such as DMSO solubility are going to be applied. The PreADME program can be used for improving the quality of HTS and combinatorial chemical libraries. You can evaluate this program at: http://preadme.bmdrc.org, free of charge.