computeParCorrelation: Compute partial correlation network
In kwanjeeraw/metabox: metabox

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

compute the partial correlation network of entities from raw or quantified data, see details.

1	computeParCorrelation(x, xtype, internalid, coef, pval, alpha, epsilon, matrix.completion, returnas)

`x`	a data frame of raw or quantified data e.g. gene expression data, metabolite intensities. Columns are samples and rows are entities e.g. genes, proteins or compounds.
`xtype`	a string specifying a node type (default = NULL). If provided and the database is installed, node attributes will be automatically retrieved from the database. For database query, the value can be one of compound, protein, gene, rna, dna.
`internalid`	a logical value indicating whether the network nodes are neo4j ids, if TRUE (default). See `convertId` for how to convert ids. It has no effect if `xtype` = NULL or there is no database installed.
`coef`	a numeric value specifying the minimum absolute partial correlation coefficient to be included in the output (from 0 to 1, default is 0.7).
`pval`	a numeric value specifying the maximum p-value to be included in the output (default is 0.05).
`alpha`	a numeric value specifying significance level of each test used in `qpAvgNrr`.
`epsilon`	a numeric value specifying the maximum cutoff value of the non-rejection rate met by the edges that are included in the qp-graph, see `qpGraph`.
`matrix.completion`	a string specifying algorithm to employ in the matrix completion operations used in `qpPAC`
`returnas`	a string specifying output type. It can be one of dataframe, list, json. Default is dataframe.

The function wraps around the functions of qpgraph. Partial correlation coefficients, p-values and correlation directions are calculated. The partial correlation coefficients are continuous values between -1 (negative correlation) and 1 (positive correlation), with numbers close to 1 or -1, meaning very closely correlated.

list of network information with the following components:

nodes:

id = node id or node neo4j id

gid = node id or node grinn id

nodename = node id or node name

nodelabel = node type if provided

edges:

source, target = node id or node neo4j id

coef = partial correlation coefficient

pval = p-value

direction = correlation direction

type = relationship type

Return empty list if error or found nothing.

If the database is installed, node attributes will be automatically retrieved from the database. Otherwise node attributes will be the original input.

Kwanjeera W kwanich@ucdavis.edu

Castelo R. and Roverato A. (2006) A robust procedure for Gaussian graphical model search from microarray data with p larger than n. J. Mach. Learn. Res., 7:2621-2650.

Castelo R. and Roverato A. (2009) Reverse engineering molecular regulatory networks from microarray data with qp-graphs. J Comput Biol, 16(2), pp. 213-27.

qpgraph, qpAvgNrr, qpGraph, qpPAC