makeplot.pairs: Plotting parameters against each other

In rwty: R We There Yet? Visualizing MCMC Convergence in Phylogenetics

Description

Makes a plot matrix of each parameter against each other (including the topology) in your analysis. The default behaviour is to just plot the first two columns of your parameter file (after removing the column for the generation number) as well as the topological distance. This usually means that you see a pairs plot with the likelihood, the tree length, and the tree toppology. We do this because some parameter files contain so many columns that the plot matrix becomes too busy. To include parameters of your choice, use the 'parameters' argument. In this function, the topological distance is calculate from the first tree in every chain.

Usage

makeplot.pairs(chains, burnin = 0, treedist = "PD", params = NA,
  strip = 1)

Arguments

chains	A list of rwty.chain objects.
burnin	The number of trees to omit as burnin.
treedist	the type of tree distance metric to use, can be 'PD' for path distance or 'RF' for Robinson Foulds distance
params	'NA', 'all', or a vector of column names to include in the plot. 'NA' gives the default behaviour (see above). 'all' plots all columns (watch out!). Choose specific columns by name with a vector.
strip	Number indicating which column to strip off (i.e., strip=1 removes first column, which is necessary for most MCMC outputs in which the first column is just the generation).

Value

pairs.plot Returns a ggplot object.

Examples

## Not run: 
data(salamanders)
makeplot.pairs(salamanders[1], burnin=20)

# plot all the variables
makeplot.pairs(salamanders[1], burnin=20, params = 'all')

# plot specific the variables (note: you always get the topological distance)
makeplot.pairs(salamanders[1], burnin=20, params = c('pi.A.', 'pi.C.', 'pi.G.', 'pi.T.'))

## End(Not run)

rwty documentation built on May 2, 2019, 4 p.m.