We calculate a "similarity score" between a pattern and the similarity group of sequences from which it was derived and obtain the averageand the standard deviation. We use the entire database as the other group and calculate the corresponding values. The statistical significance of the difference can be formulated in terms of the Student t value as follows: :


Thetvalue has a very simple meaning grapic meaning: the separation of the s (score) distribution between the two groups (In chromatography one uses an identical expression for calculating peak separation). The t value is directly applicable for optimizing patterns. If we want to describe the value in a publication, one has to look up the significance level in the corresponding Student table.