cluster: cluster

In sjessa/chromswitch: An R package to detect chromatin state switches from epigenomic data

cluster

Description

Given a sample-by-feature matrix and sample-associated metadata including their biological condition groupings, cluster samples hierarchically and use external cluster validity measures (Adjusted Rand Index, Normalized Mutual Information, and V measure) to assess the agreement between the inferred clusters and the biological conditions. Optionally, produce a heatmap reflecting the hierarchical clustering result.

Usage

cluster(ft_mat, metadata, query, heatmap = FALSE, title = NULL,
  outdir = NULL, optimal_clusters = TRUE, n_features = FALSE,
  estimate_state = FALSE, method = NULL, test_condition = NULL,
  signal_col = NULL, mark = NULL)

Arguments

ft_mat	matrix where columns are features and rows are samples as returned by summarizePeaks or binarizePeaks
metadata	A dataframe with a column "Sample" which stores the sample identifiers, and a column "Condition", which stores the biological condition labels of the samples
query	GRanges object specifying the query region
heatmap	(Optional) Logical value indicating whether to plot the heatmap for hierarchical clustering. Default: FALSE
title	(Optional) If heatmap is TRUE, specify the title of the plot, which will also be used for the output file name in PDF format
outdir	(Optional) String specifying the name of the directory where PDF of heatmaps should be saved
optimal_clusters	(Optional) Logical value indicate whether to cluster samples into two groups, or to find the optimal clustering solution by choosing the set of clusters which maximizes the Average Silhouette width. Default: TRUE
n_features	(Optional) Logical value indicating whether to include a column "n_features" in the output storing the number of features in the feature matrix constructed for the region, which may be useful for understanding the behaviour of the binary strategy for constructing feature matrices. Default: FALSE
estimate_state	(Optional) Logical value indicating whether to include a column "state" in the output specifying the estimated chromatin state of a test condition. The state will be on of "ON", "OFF", or NA, where the latter results if a binary switch between the conditions is unclear. Default: FALSE.
method	(Optional) If estimate_state is TRUE, one of "summary" or "binary", specifying which method was used to construct the feature matrix in ft_mat
test_condition	(Optional) If estimate_state is TRUE, string specifying one of the two biological condtions in metadata$Condition for which to estimate chromatin state.
signal_col	(Optional) If estimate_state is TRUE, and method is "summary", string specifying the name of the column in the original peak files which corresponds to the level of enrichment in the region, e.g. fold change
mark	(Optional) If estimate_state is TRUE, and method is "summary",string specifying the name of the mark for which ft_mat was constructed

Value

A dataframe with the region, the number of clusters inferred, the cluster validity statistics, and the cluster assignments for each sample

Examples

samples <- c("E068", "E071", "E074", "E101", "E102", "E110")
bedfiles <- system.file("extdata", paste0(samples, ".H3K4me3.bed"),
package = "chromswitch")
Conditions <- c(rep("Brain", 3), rep("Other", 3))

metadata <- data.frame(Sample = samples,
    H3K4me3 = bedfiles,
    Condition = Conditions,
    stringsAsFactors = FALSE)

region <- GRanges(seqnames = "chr19",
    ranges = IRanges(start = 54924104, end = 54929104))

lpk <- retrievePeaks(H3K4me3,
    metadata = metadata,
    region = region)

ft_mat <- summarizePeaks(lpk, mark = "H3K4me3",
cols = c("qValue", "signalValue"))

cluster(ft_mat, metadata, region)

# Estimate the state of the test condition, "Brain"
cluster(ft_mat, metadata, region,
    estimate_state = TRUE,
    method = "summary",
    signal_col = "signalValue",
    mark = "H3K4me3",
    test_condition = "Brain")

sjessa/chromswitch documentation built on Feb. 4, 2024, 2:04 a.m.