TOM: Variations on Topological dissimilarity Matrix (opposite of...

Description Usage Arguments Value Author(s) References See Also Examples

Description

Compute several variations of the (opposite of the) Topological Overlap Matrix (Zhang & Horvath, 2005).

Usage

1
TOM(data, TOMType="signed", TOMDenom = "mean", squared = FALSE, TOMPCA = FALSE, rotate="varimax", use = "pairwise.complete.obs", mineigen=c(0, .7, 1) )

Arguments

data

A dataframe, observations by variables.

TOMType

String. One of "signed" and "unsigned". If signed, the signs in the correlation matrix are considered (see Langfelder, 2013).

TOMDenom

String. One of "min", "mean" and "squared". The options "min" and "mean" are described in Langfelder (2012). The option squared is a possible variation that I am still considering. It might not work properly.

squared

Logical. If TRUE the squared direct connection is considered in the numerator.

TOMPCA

Logical. If TRUE a technique that combines both TOM and PCA is used. It can be particularly effective in presence of clones that can inflate the overlap among a pair of nodes.

rotate

String. Only relevant if TOMPCA = TRUE. A rotation technique for function principal.

mineigen

Vector of numerics. Only relevant if TOMPCA = TRUE. Only the component scores with eigenvalue larger than mineigen are considered in computing TOM. If the length of mineigen is >1, then a TOM matrix will be given in output for each value in mineigen.

Value

A matrix, each element i,j including the topological dissimilarity between a pair of elements.

Author(s)

Giulio Costantini

References

Langfelder, P. (2012). Topological overlap and the case of disappearing small modules. Technical report. Langfelder, P. (2013). Signed vs. unsigned topological overlap matrix. Technical Report. Zhang, B., & Horvath, S. (2005). A general framework for weighted gene co-expression network analysis. Statistical Applications in Genetics and Molecular Biology, 4(1).

See Also

TOMdist

Examples

1
2
3
4
5
6
## Not run: 
data <-matrix(rnorm(100), ncol=10, nrow = 10)

TOM(data)

## End(Not run)

GiulioCostantini/TOMproject documentation built on May 6, 2019, 6:29 p.m.