rtCorrClust: rtCorrClust

In WMBEdmands/MetMSLine: an automated and fully integrated pipeline for rapid processing of high-resolution LC-MS metabolomic datasets.

Description

performs retention time clustering thin intra-retention time cluster correlation coefficient (spearman's Rho) clustering.

Usage

rtCorrClust(peakTable = NULL, obsNames = NULL, rtThresh = NULL,
  corrThresh = 0.6, minFeat = 2, hclustMethod = "median",
  distMeas = "euclidean", corMethod = "spearman")

Arguments

peakTable	either a data.frame, full file path as a character string to a .csv file of a peak table in the form observation (samples) in columns and variables (Mass spectral signals) in rows. If argument is not supplied a GUI file selection window will open and a .csv file can be selected.
obsNames	character vector of observation (i.e. sample/ QC/ Blank) names to identify appropriate observation (sample) columns.
rtThresh	retention time threshold for retention time clustering.
corrThresh	correlation coefficient threshold (non-parametric Spearman's Rho) to group features within a retention time cluster.
minFeat	minimum number of features with a Retention time/ correlation cluster to consider it a group (default = 2, i.e. a cluster must contain at least 2 features to be considered a group).
hclustMethod	hierarchical clustering method to hclust.vector method of fastcluster package (default = "median").
distMeas	distance measure for retention time clustering (default = "euclidean"). see hclust.vector.
corMethod	correlation method to cor (default = "spearman). Within retention clusters dissimilarity is computed as 1-correlation coefficient.

Details

The cutree function is used to identify retention time clusters based on a cut height equal to retention time (rtThresh) supplied. Within retention time cluster correlation clustering is based on 1 - correlation coefficient dissimilarity. The cut height of intra retention time cluster correlation clusters is based on a minimum correlation coefficient threshold (default = 0.6). Following retention time/ correlation cluster identification any groups containing less than a mimimum feature number are removed (minFeat default = 2), a default value for the minFeat of 2 means that only isolated single features with no relation to another feature are removed.

Value

a list containing 2 elements:

1. the original peak table with the results of retention time correlation clustering included in additional columns "rtgroup", "CorrClusts", "RtCorrClust" as a data frame and details of the most intense feature for each group. In addition for each group details of the best sample to consider for subsequent MSn fragmentation analysis based on the most top 5 most intense samples is provided. This "best" sample for MS/MS is also selected whilst concurrently minimising the number of samples to potentially reinject for targetted MSn fragmentation analysis.

2. A data frame of the weighted means and details of the most intense feature for each group. In addition for each group details of the best sample to consider for subsequent MSn fragmentation analysis based on the most top 5 most intense samples is provided. This "best" sample for MS/MS is also selected whilst concurrently minimising the number of samples to potentially reinject for targetted MSn fragmentation analysis. @seealso hclust.vector, cutree.

WMBEdmands/MetMSLine documentation built on May 9, 2019, 10:03 p.m.