Figure 1

(a) Gene Ontology (GO) annotation graph for biological processes. The network is color coded according to overall classification of different processes. (b) Regulatory network of S. Cerevisiae, including known transcription and enzymatic interactions. The nodes represent proteins which have been colored according to their position in the above GO-graph. (c) The direct distance between two nodes A and B in a GO-annotation network is the length of the shortest path between the two nodes using the breadth-first search method, disregarding directions. Since each protein could have several GO-annotations, the distance between a pair of proteins is the shortest among all possible assigned annotations. (d) the hierarchical distance between node A and B is defined from the smallest total downstream region n(A, B) of any node that include both A and B. The hierarchical distance is the normalized D(A, B) = n(A, B)/N where N is the number of GO-nodes that has a protein in the shown regulatory network. D(A,B) captures that the distance between A and B is smaller if one is below the other, than if they are on separate branches.