metID.corrNetwork: CompMS2 correlation network generation

In WMBEdmands/compMS2Miner: an automatable metabolite identification, visualization and data-sharing R package for high-resolution LC-MS datasets

Description

Uses the MS1features matched to MS2 data to generate a correlation network to view in compMS2Explorer

Usage

metID.corrNetwork(object, ...)

Arguments

object	A "compMS2" class object.
peakTable	a data.frame in the form observation (samples) in columns and variables (Mass spectral signals) in rows. The first 3 columns must consist of: EIC number or unique peak identifier. mass-to-charge ratio of peak group. median/ peak apex retention time in seconds. These columns are utilized in the final network visualization.
obsNames	character vector of observation (i.e. sample/ QC/ Blank) names to identify appropriate observation (sample) columns.
corrThresh	correlation coefficient threshold to group features within a retention time cluster. If no value is supplied the default is to estimate an optimal cut-off value based on network attributes from a series of correlation cut-off values. This is based upon the relationship between the number of nodes, edges, number of clusters and the network density (i.e. n actual edges/n potential edges) at each correlation cut-off value. A plot showing the result of this estimation will be generated.
corrMethod	character correlation method see cor for details. default "spearman".
delta	numeric maximum p-value (following multiple testing correction) above #' which the null hypothesis (no correlation) is rejected.
MTC	character Multiple Testing Correction default is "none", see p.adjust.methods for details of options. ("holm", "hochberg", "hommel", "bonferroni", "BH", "BY", "fdr", "none"). Any p-values after multiple testing correction above the value of delta will have their corresponding correlation coefficents replaced with zero.
maxNodes	numeric above the maximum nodes the function will use the large graphing algorithm of igraph. See igraph::with_lgl else the function uses the Fruchterman-Reingold layout algorithm. See igraph::with_fr
MS2only	numeric 3 options (1-3) if 1 All EICs above corrThresh returned, if 2 only non-MS2 matched EICs which are first neighbours of at least one MS2-matched MS2 are returned, if 3 only MS2-matched EICs are returned (default = 1).
minClustSize	numeric minimum number of connected nodes that is cluster size (default =3). If a cluster of nodes is less than this number then it will be removed.

Value

"compMS2" class object with an additional network graph of any peakTable features above the correlation threshold.

References

Koh Aoki, Yoshiyuki Ogata, and Daisuke Shibata Approaches for Extracting Practical Information from Gene Co-expression Networks in Plant Biology Plant Cell Physiol (2007) 48 (3): 381-390 first published online January 23, 2007 doi:10.1093/pcp/pcm013

WMBEdmands/compMS2Miner documentation built on May 9, 2019, 10:04 p.m.