To control test the prediction {X=TF} for robustness. All the TF-sites recognition Scores and one X-site Score simulating the experimental data on DNA/protein-binding alteration are inputted into the standard STATISTICA tools. For each option-pair {similarity scale (column-listed); cluster-analysis method (line-listed)}, the corresponding similarity tree is constructed, and the TF-site clustered with the highest similarity estimate to the X-site is documented within the proper cell of this area "Robustness" . If more than a quarter of the tests gives evidence that predicted pair {X=TF} is the best, this prediction could be considered as "robust".