Figure 2

Schematic representation of the DomSign pipeline. The pipeline is divided into two parts—enzyme candidate selection and EC number annotation. In the first step, specific enzyme DSs are utilized, and all proteins with DSs within this dataset are selected as potential enzyme candidates. Simultaneously, four annotation references for the EC digits at four levels are constructed as described in Figure 1. At every level, if the “specificity” of the corresponding DS-EC pair in the annotation reference is less than the user-defined threshold, the pipeline is shut down and the previously annotated EC digits form the output. If not, the pipeline continues until the fourth EC digit has been annotated. An example of the DomSign procedure to annotate a protein is shown here. Because the specificity threshold is above the specificity of the DS-EC pair at the last level, only the first three DS-EC digits are predicted, leading to final result: EC = 1.1.1.-.