| fuzzygene_confidence_matrix {FuzzyGene} | R Documentation |
fuzzygene_confidence_matrix(fuzzygene,
layout = "square",
minConfidence = 0,
genes = NULL);
Rows and columns are exactly the same as in fuzzygene_regulation_matrix (same layout, same order), only the element value changes: W[i, j] becomes the confidence of the edge j -> i, that is the prior confidence when the edge is supported by the prior network and the absolute expression correlation |Pearson| otherwise.
Weight and confidence answer two different questions ("how strong / which direction" vs "how trustworthy"), so exporting both matrices and using weight for the edge width and confidence for the edge opacity or a cutoff is the usual combination:
let w = fuzzygene::fuzzygene_regulation_matrix(model, layout = "compact");
let c = fuzzygene::fuzzygene_confidence_matrix(model, layout = "compact");
geneExpression::write.expr_matrix(w, file = "/path/weight.csv", id = "geneID");
geneExpression::write.expr_matrix(c, file = "/path/confidence.csv", id = "geneID");
a Matrix object (rows = targets, columns = regulators, non negative values)