Would anyone be interested in a set of tools for the prediction of physiochemical properties?
I've recently been using OELib to derive molecular properties such as Molar Refractivity, LogP and Polar Surface Area.  Each one being derived from SMARTS encoded atom types and thier associated atomic property contributions [1,2].
With the use of JOELib, it should be easy to incorperate this facilty into the CDK.  Does anyone think this to be a worthwhile task for me to undertake OR is anyone already working on something similar?
[1] Wildman S.A. & Crippen G.M. (1999). Prediction of Physiochemical Parameters by Atomic Contributions. J. Chem. Inf. Comput. Sci., 39, 868-873.
[2] Ertl P., Rohde B. & Selzer P. (2000). Fast Calculation of Molecular Polar Surface Area as a Sum of Fragment-Based Contributions and Its Application to the Prediction of Drug Transport Properties. J. Med. Chem., 43, 3714-3717.
Stephen P. Jelfs  BSc(hons)
Postgraduate Research Student CISRG
Department of Information Studies
University of Sheffield, S10 2TN