Hello R Users, I am interested in using R to generate quantitative structure-activity relationships (QSARs) for small molecules given a set of molecular descriptors and biological data tab-delimited or excel file. In which fist value colum of each row is biological value and rest all are its descriptors. Has anyone done this using R ? Would you be willing to share your R scripts (or ideas) for doing this with me ? I am particularly interested in R codes for optimizing variable selection as well as validation of the generated QSAR equations, since I have ~ 1500 molecular descriptors per molecule. How I can use R to find the optimal (smallest) set of descriptors to fit the data ? If you have any idea about this using other software the please also suggest me. I am a new user of R. I am trying to use R for this. Please reply me at this mail ID because I have not joined R help group. Thanking you -- Nitish Kumar Mishra SRF, BIC, IMTECH, Chandigarh, India E-Mail Address: nitish_km at yahoo.com nitish at imtech.res.in