HPDplotBygeneBygroup {MCMC.qpcr}

Plots qPCR analysis results for individual genes

Description

For a specified gene, makes overlayed plots such as produced by HPDplotBygene()

Usage

HPDplotBygeneBygroup(model, gene, group1, group2, group3 = NULL, 
interval = "ci", colors = c("coral", "cyan3", "grey50"), 
symbols = c(19, 17, 15), jitter = 0.16, yscale = "log2", ...)

Arguments

model

model object produced by mcmc.qpcr()

gene

name of the gene to plot

group1

Combination of factors defining the first group (see HPDplotBygene() for details).

group2

Combination of factors defining the second group.

group3

(optional) Combination of factors defining the third group.

interval

'ci' (default) will plot 95% credible limits of the posterior distribution, 'sd' will plot the mean plus/minus one standard deviation of the posterior.

colors

Colors to use for different groups (see ?par -> col).

symbols

Symbols to use for different groups (see ?par -> pch).

jitter

Jitter distance between groups.

yscale

Scale on which to represent the data. In all mcmc.qpcr models the model scale is natural logarithm, which I prefer to translate into log2 or log10 (if the differences are orders of magnitude) for better human readability. The default is 'log2'; other options are 'log10' and 'native' (no rescaling of the model data). There is also a beta-option 'proportion', which is not useful for qPCR. It was added to cannibalize HPDplotBygene function for plotting results of the model operating with arcsin-square root transfromed proportions. With yscale="proportions", the plot will be on the original proportion scale but the tukey-like differences will still be reported on the asin(sqrt()) transformed scale.

...

additional parameters for HPDplotBygene() function, such as pval (see HPDplotBygene() help)

Value

Generates a point-whiskers plot, lists pairwise mean differenes between all conditions, calculates and lists pairwise p-values (not corrected for multiple testing).

Author(s)

Mikhal V. Matz, UT Austin <[e-mail]>

References

Matz MV, Wright RM, Scott JG (2013) No Control Genes Required: Bayesian Analysis of qRT-PCR Data. PLoS ONE 8(8): e71448. doi:10.1371/journal.pone.0071448