diagnostic.mcmc {MCMC.qpcr}

Plots three diagnostic plots to check the validity of the assumptions of linear model analysis.

Description

Predicted vs observed plot tests for linearity, Scale-location plot tests for homoscedasticity, and Normal QQ plot tests for normality of the residuals.

Usage

diagnostic.mcmc(model, ...)

Arguments

model

MCMCglmm object (a model fitted by mcmc.qpcr or mcmc.qpcr.gauss), obtained with additional options, 'pl=T, pr=T'

...

Various plot() options to modify color, shape and size of the plotteed points.

Value

A plot with three panels.

Author(s)

Mikhail 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

Examples

# loading Cq data and amplification efficiencies
data(coral.stress) 
data(amp.eff) 
# extracting a subset of data 
cs.short=subset(coral.stress, timepoint=="one")

genecolumns=c(5,6,16,17) # specifying columns corresponding to genes of interest
conditions=c(1:4) # specifying columns containing factors  

# calculating molecule counts and reformatting:
dd=cq2counts(data=cs.short,genecols=genecolumns,
condcols=conditions,effic=amp.eff,Cq1=37) 

# fitting the model
mm=mcmc.qpcr(
	fixed="condition",
	data=dd,
	controls=c("nd5","rpl11"),
	pr=TRUE,pl=TRUE, # these flags are necessary for diagnostics
	nitt=4000 # remove this line when analyzing real data!
)
diagnostic.mcmc(mm)