plot.rankedPartitionedRAD: Plot a 'rankedPartitionedRAD' object

Description Usage Arguments Details Value Author(s) References See Also

View source: R/plot.rankedPartitionedRAD.R

Description

This function creates a plot with 1-3 panels describing the number of loci supporting, disfavoring, or ambiguous for each tree, plotted against log-likelihood

Usage

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
## S3 method for class 'rankedPartitionedRAD'
plot(x, 
     tree.lnL.file = NULL,
     fileprefix = NULL, 
     lnL.break = NULL,
     regression = NULL,
     ci = NULL,
     widthScalar = 0.85, 
     panels = c("bestMat", "worstMat", "doubleCountMat"), 
     squareSize = switch(as.character(length(panels)), '2' = 5, '3' = 3), 
     primeTreeColor = "red", 
     primeTreeCharacter = 19, 
     highlight.points = NULL, 
     highlight.colors = NULL,
     filebase = 'DEFAULT',
     ...)

Arguments

x

an object of class rankedPartitionedRAD

tree.lnL.file

file name of the info file written by RAxML; optional

fileprefix

prefix of the files to save; as written, you have to provide some prefix to identify the files if you want a pdf, but the remainder of the file name defaults to a combination of parameters and date

lnL.break

a numeric vector, which should be the length of panels; if not null, the lowest tree likelihood to show on each panel

regression

a boolean vector, which should be the length of panels; if not null, whether to draw a linear fit line on each panel

ci

a numeric vector, the length of panels; what CI to show on each regression, or 0 for no CI

widthScalar

a multiplier on the width of the entire plot to get the plotting panels square or not as you prefer

panels

a character vector describing the order of panels; can be any of "bestMat", "worstMat", "doubleCountMat" in any order

squareSize

the size of the panels in inches; defaults to 5" for a 2-panel plot, 3" for a 3-panel plot, only because that works well in my trials

primeTreeColor

the color of the dot representing the first tree in the matrix, which should be your original tree (unpermuted) if you've been following the workflow from genTrees through this final and exciting plot

primeTreeCharacter

the character of the dot representing the first tree in the matrix

highlight.points

number of any trees to highlight

highlight.colors

colors vector or single color for highlighted points

filebase

the base of the file name for exported plots

...

additional arguments passed to plot

Details

The x-axis will either be the tree likelihoods estimated by RAxML in a separate analysis (e.g., using the entire dataset), or the summed locus likelihoods from the partitioned RAD analysis if no filename is provided. This function as written prints to pdf if you enter a fileprefix, to the screen if you don't. Default filebase pastes rad.partitioned with the date and the parameter values used. If default filebase is not desired, enter something other than "DEFAULT."

Value

This function is called primarily for its side-effect of printing plots to a file. It returns, invisibly, a list of the fit objects for each panel, only relevant if you set any elements of regression to TRUE.

Author(s)

Andrew Hipp

References

Hipp A.L., Eaton D.A.R., Cavender-Bares J., Fitzek E., Nipper R. & Manos P.S. (Accepted pending revision). A framework phylogeny of the American oak clade based on sequenced RAD data. PLoS ONE.

See Also

rank.partitionedRAD, plot.locus.dist


RADami documentation built on May 30, 2017, 8:23 a.m.