plot_CoGAPS_pattern_cor_heatmap: Plot CoGAPS pattern correlation heatmap
In jacobheng/cellwrangler: Supplemental toolkit for single-cell RNAseq analysis

Description Usage Arguments Value Examples

View source: R/plot_CoGAPS_pattern_cor_heatmap.R

plot_CoGAPS_pattern_cor_heatmap() plots a heatmap showing the correlation of a CoGAPS pattern set with a grouping variable

plot_CoGAPS_pattern_cor_heatmap(CoGAPS_res_set = NULL, Pattern_set,
  pattern_subset = NULL, annotation, group_column = NULL, group_vector,
  cluster_groups = F, order_subgroups = F,
  cluster_groups_distance = "correlation",
  cluster_groups_method = "complete", cluster_patterns = F,
  cluster_patterns_vector = NULL,
  cluster_patterns_distance = "correlation",
  cluster_patterns_method = "complete", limits = NULL,
  square_tiles = T)

`CoGAPS_res_set`	a CoGAPS result set. If set to NULL, a Pmeans matrix may be supplied directly as input for the Pattern_set parameter.
`Pattern_set`	a string specifying the name of a pattern set (e.g. "nP20") in a CoGAPS_res_set. Alternatively, if CoGAPS_res_set is set to NULL, a Pmeans matrix may be supplied directly.
`pattern_subset`	a vector specifying a subset of the pattern set to be plotted. Defaults to NULL.
`annotation`	an annotation dataframe. If group_column is set to NULL, must contain logical columns with column names corresponding to values in group_vector. Each logical column (1 or 0) should indicate whether an annotation row is of the value (column name).
`group_column`	column name of column in annotation object with values corresponding to those in group_vector to use for computing logical columns. Defaults to NULL
`group_vector`	a vector of with pre-specified values corresponding to those in group_column in the annotation object.
`cluster_groups`	logical; if groups should be clustered.
`order_subgroups`	logical; if subgroups should be ordered according to clustering of first subgroup. Subgroups can be specified in a string with separation by " - ".
`cluster_groups_distance`	distance measure used in clustering groups. Possible values are "correlation", "euclidean", "maximum", "manhattan", "canberra", "binary", or "minkowski" (defined in dist() function.)
`cluster_groups_method`	clustering method used for groups. Accepts the same values as hclust: "ward.D", "ward.D2", "single", "complete", "average" (=UPGMA), "mcquitty" (=WPGMA), "median" (=WPGMC) or "centroid" (=UPGMC).
`cluster_patterns`	logical; if patterns should be clustered.
`cluster_patterns_vector`	a vector specifying the order of patterns. If NULL, patterns will be clustered according to distance and method specified below.
`cluster_patterns_distance`	distance measure used in clustering patterns. Possible values are the same as cluster_groups_distance().
`cluster_patterns_method`	clustering method used for patterns. Possible values are the same as cluster_groups_method().
`limits`	limits of the color scale for the heatmap.
`square_tiles`	logical; if tiles should be made square

a ggplot object

plot_CoGAPS_pattern_cor_heatmap(CoGAPS_res_set = my_CoGAPS_res, Pattern_set = "nP30",annotation=pData(cds), 
group_column=NULL, group_vector=genotype_vector, cluster_groups=T, order_subgroups=T, 
cluster_groups_distance = "correlation", cluster_groups_method = "complete", cluster_patterns= T, 
cluster_patterns_distance = "correlation", cluster_patterns_method = "complete")