SRC's LogKow (KowWin) Program

The LogKow (KowWin) program estimates the log octanol/water partition coefficient (log P) of organic chemicals using an atom/fragment contribution method developed at SRC. KowWin runs under Windows (Windows 3.1, 95, 98, NT) making the estimation of octanol/water partition coefficients more convenient and accurate.

Try our Online Version of LogKow to examine the program's estimation ability and experimental database retrieval.

The KowWin Program methodology is described in the following journal article: Meylan, W.M. and P.H. Howard. 1995. Atom/fragment contribution method for estimating octanol-water partition coefficients. J. Pharm. Sci. 84: 83-92.

The octanol-water partition coefficient is a physical property used extensively to describe a chemical's lipophilic or hydrophobic properties. It is the ratio of a chemical's concentration in the octanol-phase to its concentration in the aqueous phase of a two-phase system at equilibrium. Since measured values range from <10-4 to >10+8 (at least 12 orders of magnitude), the logarithm (log P) is commonly used to characterize its value. Log P is a valuable parameter in numerous quantitative structure-activity relationships (QSAR) that have been developed for the pharmaceutical, environmental, biochemical and toxicological sciences. For example, two additional estimation programs available from Syracuse Research Corporation (WsKow and DermWin) use KowWin’s estimation methodology and experimental log P database retrieval system to predict water solubilities and dermal permeability coefficients, respectively.

Methods for chemical structure input now support SMILES notations as well as chemical structures produced in popular chemical drawing programs. Enhanced features of the Windows version include batch-mode data entry and an expanded database of experimental values. The KowWin program also has a unique feature to allow the user to adjust the estimation using an experimental octanol/water partition coefficient from a structurally-related chemical compound.

Example Entry/Results Windows

KowWin includes the following program features:

(1) A "Structure" window that draws the input SMILES notation. Users can see the structure that's being estimated.

(2) Structures in the "Structure" window can be printed, copied to the clipboard, and saved as MDL Information Systems MOL files or ISIS/Draw SKC files.

(3) Structures in scientific file formats other than SMILES notations can be imported directly into KowWin. KowWin converts these other formats into SMILES and automatically places the conversion in the entry field.

(4) Batch-Mode calculation of text files of SMILES strings (no file size limit).

(5) Batch-Mode calculation of text files containing lists of CAS numbers (no file size limit).

(6) KowWin can use MDL Information System's SDFile format to run batches (no SD File size limit).

(7) KowWin v1.6 adds a new feature that allows users to automatically generate zwitterionic log P estimates from various non-zwitterionic SMILES.

(8) Optional Experimental Value Adjustment (EVA) estimation methodology.

Using a training set of 2,410 chemicals with 175 groups and 280 correction factors, a correlation coefficient of 0.98 was found between the experimental and estimated log Kow values. The method has been evaluated using a validation set of experimental values for 10,550 chemicals other than those included in the training set with a resulting correlation coefficient of 0.94 and a mean error of 0.35. Additional information concerning KowWin's accuracy is illustrated below.

Estimation Accuracy of LogKow Program

The figures below illustrate the LogKow program's ability to estimate accurate log P values. They were generated using our complete database (13,058 compounds) of reliable, experimental log P values. For the entire dataset, KowWin estimates had a standard deviation of 0.436, an absolute mean error (absolute deviation) of 0.316, and a correlation coefficient (r^2) of 0.95.

ScatterPlot of Estimated vs. Experimental Log P

Histogram of LogKow Estimation Error