RNAseqNet (version 0.1.5)

GLMnetToGraph: Convert the result of imputedGLMnetwork or a matrix into a network.

Description

GLMnetToGraph combines the m inferred networks, obtained from m imputed datasets, into a single stable network or convert a matrix of coefficients of a GLM model into a network (non zero coefficients are converted to edges)

Usage

GLMnetToGraph(object, threshold = 0.9)

Value

an 'igraph' object. See igraph

Arguments

object

an object of class HDpath as obtained from the function imputedGLMnetwork or a squared matrix with zero and non zero values

threshold

the percentage of times, among the m imputed networks, that an edge has to be predicted to be in the final network. Used only for objects of class HDpath. Default to 0.9

Author

Alyssa Imbert, Nathalie Vialaneix,

References

Imbert, A., Valsesia, A., Le Gall, C., Armenise, C., Lefebvre, G. Gourraud, P.A., Viguerie, N. and Villa-Vialaneix, N. (2018) Multiple hot-deck imputation for network inference from RNA sequencing data. Bioinformatics. tools:::Rd_expr_doi("10.1093/bioinformatics/btx819").

See Also

imputedGLMnetwork, igraph

Examples

data(lung)
data(thyroid)
nobs <- nrow(lung)
miss_ind <- sample(1:nobs, round(0.2 * nobs), replace = FALSE)
lung[miss_ind, ] <- NA
lung <- na.omit(lung)
lambdas <- 4 * 10^(seq(0, -2, length = 10))
if (FALSE) {
lung_hdmi <- imputedGLMnetwork(lung, thyroid, sigma = 2, lambdas = lambdas,
                               m = 10, B = 5)
lung_net <- GLMnetToGraph(lung_hdmi, 0.75)
lung_net
plot(lung_net)
}