Figure 2From: Why is Tanimoto index an appropriate choice for fingerprint-based similarity calculations?Scheme of the data generation. The SRD procedure was repeated 1000 times to eliminate the effect of random choices. Sum of ranking differences was calculated for 1000 data sets and gathered in an output file. The final output file contains a table with all of the SRD values for each similarity measure (n) on every dataset (m).Back to article page