Table 2.

Summary of the procedures proposed to infer putative missing interactions in data-based network models.

Procedure	Application to inferring putative interactions
PROCEDURE 1
Adding missing links in congruence with available experimental data that can be represented in single rows of truth tables	This is probably the most simple procedure. It allows modifying the network adding missing putative interactions, and at the same time the regulatory effects of the nodes whose role is based on experimental that is represented by single rows of the true tables is preserved. Examples of experiments represented by single rows are loss and gain-of-function mutants
PROCEDURE 2
Adding missing links while maintaining the sign of the regulation	Prevents changes in the regulatory sign of genes when we introduce putative missing interactions
PROCEDURE 3
Adding missing links while maintaining documented protein–protein interactions	This procedure guarantee that the joint action of proteins acting as complexes is respected in the new rows that result from the addition of new interactions. However, it allows new complexes to be formed, replacing, deleting or including one or more components in the complex
PROCEDURE 4
Adding missing links while maintaining necessary protein–protein interactions	Procedure 4 is similar to procedure 3, since it also guarantee the joint action of proteins acting as complexes. However, this procedure do not allows the substitution or deletion of any of the components of the complexes. Importantly, it does allow the incorporation of other components in the complex
PROCEDURE 5
Adding missing links without independent TGEN activity	Procedure 5 prevents the generation of BFs where one or more regulator has no effect on its target gene
PROCEDURE 6
Adding missing links without ambiguous regulators	Procedure 6 prevents the emergence of nodes that act as global positive and negative (ambiguous) regulators at the same time