CoExpNet uses a multivariate Gaussian model to infer the gene interaction networks based on large-scale gene expression profile. CoExpNet gives a complete network with scores on the interactions. The scores show the strength of the interaction. The input of CoExpNet is a tab-delimited file (i.e. expression profile) in which each column represents a gene and each row a sample (see the sample input). To have a reasonable network, there usually need to be at least five samples per each gene. ‘Lambda’ is a parameter to prevent the over-fitting. Usually values between (0.1,0.001) works well; and ‘ratio’ identifies the fraction of interactions that we are interested in and seems significant. The output is a file in which each line shows an interaction between two genes and their score.

Opgen-Rhein, Rainer, and Korbinian Strimmer. "From correlation to causation networks: a simple approximate learning algorithm and its application to high-dimensional plant gene expression data." BMC systems biology 1.1 (2007): 37.