File with alignment probabilities produced by parseAlignment
outFile
Prefix for the output files.
outputType
Output type, possible values: theta, RPKM, counts. This is only relevant when the samples option is used. Default: theta.
trInfoFile
File containing transcript information. (Necessary for RPKM output)
seed
Sets the initial random seed for repeatable experiments.
samples
Number of samples to be generated from the posterior distribution. Default: no samples are generated.
verbose
Verbose output.
veryVerbose
Very verbose output.
procN
Maximum number of threads to be used. The program will not use more threads that there are MCMC chains.
Advanced options:
optLimit
The optimisation limit in terms of minimal gradient or change of bound.
optMethod
The optimisation method, use "FR", "HR", or "steepest".
pretend
Do not execute, only print out command line calls for the C++ version of the program.
Details
This function runs Variational Bayes algorithm to estimate the transcript expression.
The input is the .prob file containing alignment probabilities which were produced by parseAlignment.
Other optional input is the transcript information file specified by trInfoFile and again produced by parseAlignment.
It is much faster inference than MCMC which estimates mean expression equally well.
However, the posterior is in form of Dirichlet distribution with underestimated variance.
Use this method in cases when you are only interested in mean expression.
Value
.m_alphas
file containing mean relative expression of transcripts theta and parameters of the Dirichlet distribution. Please note the first line in the file corresponds to the noise transcript.
If option samples is used, the program also generates samples based of the outputType, the default would be file with extension ".VBtheta".