plotPriors: Plots the density of the log values estimated for the mean...

In baySeq: Empirical Bayesian analysis of patterns of differential expression in count data

Description

This function plots the density of the log values estimated for the mean rate in the data used to estimate a prior distribution for data under the assumption of a Negative Binomial distribution. This function is useful for looking for bimodality of the distributions, and thus determining whether we should try and identify data with no true expression.

Usage

plotPriors(cD, group, par = 1)

Arguments

cD	countData object, for which priors have been estimated using the assumption of a Negative Binomial distribution (see getPriors.NB).
group	Which group should we plot the priors for? In general, should be the group that defines non-differentially expressed data. Can be defined either as the number of the element in 'cD@groups' or as a string which will be partially matched to the names of the 'cD@groups' elements.
par	The parameter of the prior that will be plotted.

Details

If the plot of the data appears bimodal, then it may be sensible to try and look for data with no true expression by using the option nullPosts = TRUE in getLikelihoods.NB.

Value

Plotting function.

Author(s)

Thomas J. Hardcastle

See Also

getPriors.NB, getLikelihoods.NB

Examples

# We load in a `countData' object containing the estimated priors (see `getPriors').

data(CDPriors)

plotPriors(CDPriors, group = "NDE", par = 1)

baySeq documentation built on Nov. 8, 2020, 5:43 p.m.