ScISI (version 1.44.0)

mergeBGMat: A function that merges two bipartite graph (BG) incidence matrices into one.

Description

This function takes the union of the row names of mat1 and mat2 for the row names of the aggregate matrix, and takes the union of the complexes of mat1 and mat2. The resulting matrix is also an incidence matrix so an entry of unity impiles protein p is a member of complex C.

Usage

mergeBGMat(mat1, mat2, toBeRm)

Arguments

mat1

The first bipartite graph incidence matrix

mat2

The second bipartite graph incidence matrix

toBeRm

A character vector of complexes to be removed

Value

An aggregate bipartite graph incidence matrix.

Details

This function takes two bipartite graph matrices and merges them into one aggregate incidence matrix where informational redundancy is removed.

The rows of the aggregate matrix is indexed by the union of the rownames of mat1 and mat2. It is important that the rownames of mat1 and the rownames of mat2 are from the same name set (e.g. for yeast, only the standard gene names should be used).

The columns will be indexed by different protein complexes. If two protein complexes are identical, say C-i = K-j, then either C-i or K-j should be listed in toBeRm argument (given as an argument). When the matrices are merged, only one of the two will be kept. The vector toBeRm is generated by calling either runCompareComplex or findSubComp.

Examples

#go = getGOInfo()
#mips = getMipsInfo()
#goM = createGOMatrix(go)
#mipsM = createMipsMatrix(mips)
#cc = runCompareComplex(mipsM, goM, byWhich = "ROW")
#merged = mergeBGMat(mipsM, goM, cc$toBeRm)