Composite Plot

Generate a Composite Plot PNG from composite data like the output in TagPileup.

When reviewing the output from several TagPileup runs, this tool is convenient for the bulk visualization of these composite data.

File input (TSV)

The tool expects a text file with any extension formatted like the TagPileup's composite output of tab-delimited values with a single header column and the first row defining the domain:

<blank>	-499	-498	-497	...	499	500
Sample1_read1_sense.cdt	0.0	0.13333333333333333	0.0	0.0	...	0.0	0.06666666666666667
Sample1_read1_anti.cdt	0.06666666666666667	0.0	0.0	...	0.26666666666666666	0.0

The format allows for any number of rows, but the same number of columns.

caution

Differing number of columns between rows will throw an error.

Include Legend

You may choose to include a legend for each heatmap which will specify which color was used for each composite vector, labeled by the row header in the first column.

Image dimensions

The image height and width specify the number of pixels to squish or expand the final output .png heat map image. Image dimensions are enabled with the selection of "Output Image."

Output options (PNG)

"Output image" must be selected to save the output image in the GUI but either way, the composite image will pop up for the user in a new window when run in the graphical interface. If you choose to output the image, it will be written to a filename derived from the input with its extension stripped and the _plot.png suffix appended.

Command Line Interface

Usage:

java -jar ScriptManager.jar figure-generation composite-plot [-hlV] [-o=<output>]
[-t=<title>] [-x=<pixelWidth>] [-y=<pixelHeight>] [-c=<colors>...]...
<compositeData>

Output Options

Option	Description
-o, --output=<output>	specify output filename, please use PNG extension (default=Input filename with "_compositePlot.png" appended to the name in working directory of ScriptManager

Plot Design Options

Option	Description
-t, --title=<title>	set title (default uses input file name)
-l, --legend	add a legend (default=no legend)
-x, --width=<pixelWidth>	indicate a pixel width for the plot (default=500)
-y, --height=<pixelHeight>	indicate a pixel height for the plot (default=270)
-c, --custom-colors=<colors>...	indicate colors to use for each series. Must indicate a number of colors that matches number of dataseries default behavior: if one series input, use black if two series input, use blue(sense) and red(anti) if greater than two series, cycle through a set of 20 preset colors.