romicsVariableHclust: romicsVariableHclust()

In PNNL-Comp-Mass-Spec/RomicsProcessor: omics-oriented package for reproducible and sharable data processing pipelines

romicsVariableHclust()

Description

Plots a hierarchical clustering for the variables and adds two columns in the statistical layer of the romics_object indicating the order of the clustering and the clusters identifier of each variable.

Usage

romicsVariableHclust(
  romics_object,
  clusters = 8,
  method_dist = "euclidean",
  method_hclust = "ward.D",
  plot = TRUE,
  scale_data = TRUE,
  ANOVA_filter = "none",
  p = 0.05,
  statCol = "none",
  statCol_filter = "<=0.05",
  statCol2 = "none",
  statCol2_filter = "<=0.05"
)

Arguments

romics_object	A romics_object created using romicsCreateObject().
clusters	Numerical of length 1. Indicates the number of clusters to be used for the coloring of the variable hclust. 8 by default.
method_dist	Dist method to be used. This must be one of "euclidean", "maximum", "manhattan", "canberra", "binary", "minkowski", "pearson", "spearman" or "kendall".
method_hclust	Agglomeration method to be used. This should be (an unambiguous abbreviation of) one of "ward", "single", "complete", "average", "mcquitty", "median" or "centroid".
plot	Boolean indicating wether the clustering should be plotted or not.
scale_data	Boolean indicating wether the data should be scaled or not.
ANOVA_filter	Either 'none', 'p' or 'padj'. Indicates if an the ANOVA filter has to be used to plot the Heatmap (only the features below the filter will be displayed on the heatmap)
p	Numerical of length 1 indicating the value of the ANOVA_filter cutoff (anything below this value will be conserved).
statCol	A column contained in the statistical layer of the romics_object, the list of columns can be obtained by using the function romicsCalculatedStats().
statCol_filter	Character to indicate how this column should be filtered (e.g. '<=0.05','>0.05','==1', '==TRUE', '>2')
statCol2	A column contained in the statistical layer of the romics_object, the list of columns can be obtained by using the function romicsCalculatedStats().
statCol2_filter	Character to indicate how this column should be filtered (e.g. '<=0.05','>0.05','==1', '==TRUE', '>2')

Details

Create a customizable and filterable hierarchical clustering based on the romics_object statistics Layer. The ANOVA filter enables to restrict the variable displayed to be only the ones passing an ANOVA

2 stat column filters (StatCol) can be set simultaneously to restrict the variable displayed. Each filter enable to sort based on a given column of the statistics layer (statCol_filter) of an romics_object (the list of columns can be obtained by using the function romicsCalculatedStats()) using a specific text (statCol_text) this text indicate what parameter should be used to filter this column (example: column has to be positive -> statCol_text= '>0'). Note that the ANOVA filter is applied first (if any) and then the filters are applied sequencially (first, then second, then third).

Value

return an Romics_object containing a new columns in its statistical layer : hclust_cluster. If this column was pre-existing it will be replaced

Author(s)

Geremy Clair

PNNL-Comp-Mass-Spec/RomicsProcessor documentation built on March 18, 2023, 5:14 a.m.