runMarker: Run identification of unique biomarkers
In xlucpu/MOVICS: Multi-Omics integration and VIsualization in Cancer Subtyping

Description Usage Arguments Value References Examples

This function aims to identify uniquely and significantly expressed (overexpressed or downexpressed) biomarkers for each subtype identified by multi-omics integrative clustering algorithms. Top markers will be chosen to generate a template so as to run nearest template prediction for subtype verification.

runMarker(
  moic.res = NULL,
  dea.method = c("deseq2", "edger", "limma"),
  prefix = NULL,
  dat.path = getwd(),
  res.path = getwd(),
  p.cutoff = 0.05,
  p.adj.cutoff = 0.05,
  dirct = "up",
  n.marker = 200,
  doplot = TRUE,
  norm.expr = NULL,
  annCol = NULL,
  annColors = NULL,
  clust.col = c("#2EC4B6", "#E71D36", "#FF9F1C", "#BDD5EA", "#FFA5AB", "#011627",
    "#023E8A", "#9D4EDD"),
  halfwidth = 3,
  centerFlag = TRUE,
  scaleFlag = TRUE,
  show_rownames = FALSE,
  show_colnames = FALSE,
  color = c("#5bc0eb", "black", "#ECE700"),
  fig.path = getwd(),
  fig.name = NULL,
  width = 8,
  height = 8,
  ...
)

`moic.res`	An object returned by 'getMOIC()' with one specified algorithm or 'get%algorithm_name%' or 'getConsensusMOIC()' with a list of multiple algorithms.
`dea.method`	A string value to indicate the algorithm for differential expression analysis. Allowed value contains c('deseq2', 'edger', 'limma').
`prefix`	A string value to indicate the prefix of differential expression file (use for searching files).
`dat.path`	A string value to indicate the path for saving the files of differential expression analysis.
`res.path`	A string value to indicate the path for saving the results for identifying subtype-specific markers.
`p.cutoff`	A numeric value to indicate the nominal p value for identifying significant markers; pvalue < 0.05 by default.
`p.adj.cutoff`	A numeric value to indicate the adjusted p value for identifying significant markers; padj < 0.05 by default.
`dirct`	A string value to indicate the direction of identifying significant marker. Allowed values contain c('up', 'down'); 'up' means up-regulated marker, and 'down' means down-regulated marker.
`n.marker`	A integer value to indicate how many top markers sorted by log2fc should be identified for each subtype; 200 by default.
`doplot`	A logic value to indicate if generating heatmap by using subtype-specific markers. TRUE by default.
`norm.expr`	A matrix of normalized expression data with rows for genes and columns for samples; FPKM or TPM without log2 transformation is recommended.
`annCol`	A data.frame storing annotation information for samples.
`annColors`	A list of string vectors for colors matched with annCol.
`clust.col`	A string vector storing colors for annotating each subtype at the top of heatmap.
`halfwidth`	A numeric vector to assign marginal cutoff for truncating values in data; 3 by default.
`centerFlag`	A logical vector to indicate if expression data should be centered; TRUE by default.
`scaleFlag`	A logical vector to indicate if expression data should be scaled; TRUE by default.
`show_rownames`	A logic value to indicate if showing rownames (feature names) in heatmap; FALSE by default.
`show_colnames`	A logic value to indicate if showing colnames (sample ID) in heatmap; FALSE by default.
`color`	A string vector storing colors for heatmap.
`fig.path`	A string value to indicate the output path for storing the marker heatmap.
`fig.name`	A string value to indicate the name of the marker heatmap.
`width`	A numeric value to indicate the width of output figure.
`height`	A numeric value to indicate the height of output figure.
`...`	Additional parameters pass to pheatmap.

A figure of subtype-specific marker heatmap (.pdf) if doPlot = TRUE and a list with the following components:

unqlist a string vector storing the unique marker across all subtypes.

templates a data.frame storing the the template information for nearest template prediction, which is used for verification in external cohort.

dirct a string value indicating the direction for identifying subtype-specific markers.

heatmap a complexheatmap object.

Gu Z, Eils R, Schlesner M (2016). Complex heatmaps reveal patterns and correlations in multidimensional genomic data. Bioinformatics, 32(18):2847-2849.