Dear Colleagues : I have a rigourous and efficient method to compare molecular similarity (of small/medium) molecules. It has been benchmarked for a small test data set. Does someone have a set of 10-15 molecules (typically upto 40 atoms each) for which molecular similarity has been worked out by some o method and is also available through some activity/property data? I will appreciate receiving such test data sets. Thanks..................................Shridhar Gadre