SIM (version 1.42.0)

sim.plot.zscore.heatmap: Association heatmap from z-scores

Description

Produces an association heatmap that shows the association (standardized influence) of each independent feature (expression measurement) with each dependent feature (copy number measurement). A P-value bar on the left indicates test signficance. A color bar on top indicates genes with mean z-scores across the signficant copy number probes above a set threshold. A summary of the copy number data helps to identify what copy number alterations are present in a region of association with expression. Positive association can mean copy number gain and increased expression, or deletion and decreased expression. The heatmaps can also be used in an exploratory analysis, looking for very local effects of copy number changes (usually small amplifications) on gene expression, that do not lead to a significant test result.

Usage

sim.plot.zscore.heatmap(input.regions = "all chrs",  		input.region.indep = NULL, 
		method = c("full", "smooth", "window", "overlap"),
		adjust = ~1, 
		significance = 0.2, 
		z.threshold = 3, 
		colRamp = colorRampPalette(c("red", "black", "green")), 
		add.colRamp = colorRampPalette(c("blue", "black", "yellow"))(7),
		show.names.indep = FALSE, 
		show.names.dep = FALSE,
		adjust.method = "BY", 
		scale, 
		add.scale,
		add.plot = c("smooth", "none", "heatmap"), 
		smooth.lambda = 2,		
		pdf = TRUE,		
		run.name = "analysis_results",...)

Arguments

input.regions

vector indicating the dependent regions to be analyzed. Can be defined in four ways: 1) predefined input region: insert a predefined input region, choices are: “all chrs”, “all chrs auto”, “all arms”, “all arms auto” In the predefined regions “all arms” and “all arms auto” the arms 13p, 14p, 15p, 21p and 22p are left out, because in most studies there are no or few probes in these regions. To include them, just make your own vector of arms. 2) whole chromosome(s): insert a single chromosome or a list of chromosomes as a vector: c(1, 2, 3). 3) chromosome arms: insert a single chromosome arm or a list of chromosome arms like c("1q", "2p", "2q"). 4) subregions of a chromosome: insert a chromosome number followed by the start and end position like "chr1:1-1000000" These regions can also be combined, e.g. c("chr1:1-1000000","2q", 3). See integrated.analysis for more information.

input.region.indep

indicating the independent region which will be analysed in combination of the dependent region. Only one input region can given using the same format as the dependent input region.

method

this must be the either full, window, overlap or smooth but the data should generated by the same method in integrated.analysis.

adjust

This variable must be a vector with the same length as samples or FALSE. The vector will be transformed to a factor and the levels of this will be coloured according to their subtype. When subtype=FALSE, all the samples will be coloured black.

significance

The threshold for selecting significant P-values.

z.threshold

Threshold to display a green or red bar in the color bar on top of the heatmap for independent features with mean z-scores above z.threshold (high positive association) or below -z.threshold (high negative association).

colRamp

Palette of colors to be used in the heatmap.

add.colRamp

Palette of colors to be used in the added plot.

show.names.indep

logical if set to TRUE, displays the names (indep.id and in dep.symb entered in the assemble.data) of the independent features with mean z-scores above or below the z.threshold in the heatmap.

show.names.dep

logical if set to TRUE, displays the names (dep.id and dep.sy mb entered in the assemble.data) of the significant dependent features in the heatmap.

adjust.method

Method used to adjust the P-values for multiple testing, see p.adjust. Default is "BY" recommended when copy number is used as dependent data. See SIM for more information about adjusting P-values.

scale

Vector specifying the color scale in the heatmap. If scale="auto", the maximum and minimum value of all z-scores will be calculated and set as the limits for all analyzed regions. Another option is to define a custom scale, e.g. scale = c(-5,5).

add.scale

Vector specifying the color scale in the left plot near the heatmap. If scale="auto", the maximum and minimum value of all the values will be calculated and set as the limits for all analyzed regions. Another option is to define a custom scale, e.g. scale = c(-5,5).

add.plot

Summary plot of copy number data in left panel. Either "smooth","heatmap", or "none". The "smooth" plot smoothes the copy number log ratios per sample, see quantsmooth for more details. The "heatmap" method produces an aCGH heatmap where green indicates gain, and red loss. The scale of the aCGH heatmap is automatically set to the min and max of the aCGH measurements of the analyzed regions. Default is plot.method = "none", no additional plot will be drawn.

smooth.lambda

Numeric value, specifying the quantile smoothing parameter for plot.method="smooth". See quantsmooth and references for more information.

pdf

logical; indicate whether to generate a pdf of the plots in the current working directory or not.

run.name

This must be the same a given to integrated.analysis

...

not used in this version

Value

No values are returned. The results are stored in a subdirectory of run.name as pdf.

Details

The sim.plot.zscore.heatmap function can only run after the integrated.analysis is run with zscores = TRUE.

The results are returned as a single-page pdf containing an association heatmap of the regions listed in input.regions. For high-density arrays large files will be produced, both demanding more memory available from your computer to produce them as well as being heavier to open on screen. To avoid this, analyze chromosome arms as units instead of chromosomes, both here and in input.regions = "all arms". The heatmap contains the z-scores generated by the function integrated.analysis with zscores=TRUE. The dependent features are plotted from bottom to top, the independent features from left to right. Positive associations are shown in green, negative associations in red (color scale on the right). At the left side of the heatmap a color bar represents the multiple testing corrected P-values of the probes in the dependent data (copy number), also with a color legend. Dependening on which plot.method is used, a summary of copy number changes is shown on the left. At the top of the heatmap is a color bar corresponding to the mean z-scores of the independent features (expression data) that are above or below the z.threshold. If show.names.indep is set to TRUE, labels will be drawn for the probes with mean z-scores greater than z.threshold or lower than -z.threshold at the bottom of the heatmap. If show.names.dep is set to TRUE, labels will be drawn for the significant dependent probes lower than significance to the right of the heatmap.

References

Eilers PH, de Menezes RX. 2005 Apr 1, Quantile smoothing of array CGH data. Bioinformatics, 21(7):1146-53. Wang P, Kim Y, Pollack J, Narasimhan B, Tibshirani R. 2005, A method for calling gains and losses in array CGH data. Biostatistics, 6 :45-58.

See Also

SIM, tabulate.pvals, tabulate.top.dep.features, tabulate.top.indep.features, sim.plot.overlapping.indep.dep.features

Examples

#first run example(assemble.data)
#and example(integrated.analysis)

#plot the zscores in a heatmap
sim.plot.zscore.heatmap(input.regions = "8q",  adjust.method = "BY", run.name = "chr8q", pdf = FALSE)

sim.plot.zscore.heatmap(input.regions = "8q", 
                        method="full", 
                        significance = 0.05,                        
                        z.threshold = 1, 
                        colRamp = colorRampPalette(c("red", "black", "green"))(15), 
                        show.names.indep=TRUE, 
                        show.names.dep=TRUE, 
                        adjust.method = "holm",  
                        add.plot = "heatmap", 
                        smooth.lambda = 2,                        
                        pdf = FALSE,                         
                        run.name = "chr8q")
                        
sim.plot.zscore.heatmap(input.regions = "8q", 
                        method="full", 
                        significance = 0.05,                        
                        z.threshold = 1,                        
                        show.names.indep = TRUE, 
                        show.names.dep = TRUE,                       
                        add.plot = "none", 
                        smooth.lambda = 2,
                        scale = c(-2, 2),                       
                        pdf = FALSE, 
                        run.name = "chr8q")