myheatmap2: Heatmap2

In NBBttest: Negative Binomial Beta t-Test

Heatmap2

Description

This function is used to display heatmap of differential expressions of genes or isoforms detected by NBBttest in the real count data.

Usage

myheatmap2(dat, IDcol=1, nci=NULL, r, colrs=
"greenred", rwcex=2.8, clcex=2.8, x=10, tree=
"both", method="euclidean", ky=1.5, rowBarColor
=NULL,  colBarColor=NULL, labrow="yes", 
labcol="yes", rsort="yes", adjrow=c(0.2, 0.0 ), 
adjcol = c(1, 1) , rwangle=30, clangle=30, 
maptitle="",keyvalue)

Arguments

dat	object of NBBttest, including information columns, data columns, tvalue column, pvalue and selection column.
nci	int value, number of columns for information of genes or isoforms, like "geneid", ""tagetid"", "straind", annotation etc.
IDcol	int value, indicating which column is ID column where IDs will be shown in heatmap rows.
r	int value, indicating column number of count data including empty columns if there are multiple datasets with the same row names.
colrs	heatmap colors. User has 8 options: "redgreen", "greenred", "redblue", "bluered", "cm.colors", "terrain.colors","topo.colors", and"heat.colors". Default colrs is "redgreen".
rwcex	numeric argument: nonnegative number, used as cex.axis for the row axis labeling. Default value is 1.8.
clcex	numeric argument: nonnegative number, used as cex.axis for the column axis labeling. Default value is 1.8.
x	numeric argument: nonnegative number, used as argument of cm.colors(x), terrain.colors(x) and topo.colors(x), Default value is 10.
tree	tree(s) drawn on row or column or both. User has four options: "both" for drawing trees in both row and column,"row" for drawing tree only in row,"column" for drawing tree in only column, and "none" for no tree specified on rows and columns. If tree =both, then colomns and rows are sorted by trees. If tree = "row", the colmns are not sorted. If tree ="column", then rows are not sorted. Default is "both".
method	method to be chosen to calculate distance between columns or rows. It has four options: "euclidean", "pearson","spearman" and "kendall". The latter three are d=1-cc where cc is correlation coefficients. Default is "euclidean".
ky	numeric argument: nonnegative number, is used to determine key size. The defalut =1.5.
rowBarColor	(optional) character vector for RowSideColors and colRow. Length of rowBarColor equals to gene or isoform number or row number. rowBarColor contains the color names of classes or typpes or groups of row names and show row color side bars and color row names. If row names, for example, genes in row are not classified, then we suggest user use its default value: rowBarColor=NULL.
colBarColor	(optional) character vector for colSideColors and colCol. Length of colBarColor equals to sample number. colBarColor contains the color names of classes or typpes or groups of coloumn names and show column color side bars and color column names. If column names, for example, samples in column are just two types, then we suggest user use its default value: colBarColor=NULL.
labrow	a string, logistic value. Rows on heatmap are labeled with genes or targets if labrow="yes", otherwise, the rows are not labeled.
labcol	a string, logistical value. Columns on heatmap are labeled with samples or treatments if labcol="yes", otherwise, the columns are not labeled.
adjrow	two numeric values. The first value used to adjust left or right position (x-axis) of labels and the second value is used to adjust up or down position (y-axis) of labels. The fadault values are c(0.3, 0.0).
adjcol	two numeric values. The first value used to adjust left or right position (x-axis) of labels, and the second value is used to adjust up or down position (y-axis) of labels. The fadault values are c(1, 1 ).
rsort	logistical value. If choose tree="both" or "row", then rsort does not work. However, if tree ="none" or "columnn", then rsort ="yes" will force rows to be sorted in descent way. The default is "yes"
rwangle	heatmap object: angle of xlab. Thed default value is 30.
clangle	heatmap object: angle of ylab. The default value is 30.
maptitle	string argument for giving heatmap title. The default value is set to be empty string ' '.
keyvalue	sting argument for name of key x-axis.

Details

This function uses selection to choose genes or isoforms in the data and then to normalize the selected data by using n-scale. Different from z-score, n-score does not follow standard normal distribution with mean = 0 and variance =1 for all rows but it has the same largest count in all rows and shows multiple colors for numeric difference between two conditions. This function has multiple options to select map color, distance, cluster and x-lab and y-lab angles. This function can be able to display multiple datasets in two ways: if multiple datasets have the same row names or features, the these datasets are put onto the different columns separated with empty column named with dataset names.If multiple datasets have the same column names of the datasets, then put them on different rows saparated with empty rows named with dataset names or whatever names user specifies.

Value

not return value but create a heatmap.

Note

requires gplots and grDevices. If the data for heatmap are multiple datasets, then tree="none" and sort="no", otherwise, myheatmap2 will get error. So before performing myheatmap2, user should sort the data in excel.

Author(s)

[e-mail]

See Also

heatmap.2, grDevices, and myheatmap.

Examples

data(result)
colclass=c("1","1","1","2","2","2")
oldpar <- par(no.readonly =TRUE)
on.exit(par(oldpar))
par(mar=c(7.5,5.5,3.5,1.2))
par(oma=c(3,1,1,3))
myheatmap2(dat=result, IDcol=1, nci=7, r=6, 
colrs="greenred", rwcex=1.8, clcex=1.8, x=10, 
tree="both", method="euclidean", ky=1.5, 
rowBarColor=NULL,  colBarColor=colclass, 
labrow="no", labcol="yes", adjrow=c(0.2, 0.0 ), 
adjcol = c(1, 1) , rwangle=0, clangle=30,
maptitle="My heatmap2",keyvalue="count")

NBBttest documentation built on May 30, 2022, 1:05 a.m.