bitseq_fitGPs: Fitting GP models for the BitSeq example

In GPrank: Gaussian Process Ranking of Multiple Time Series

Description

Function for fitting two GP models and computing Bayes factors, i.e., the ratio of the maximum marginal likelihood estimates of the two GP models, where the models are:

• null model: GP with only white noise covariance matrix

• alternative model: GP with squared exponential and white noise covariance matrices

Optionally, log Bayes factors and the parameter estimates can be written to output files whose names are specified with fileName_logBF, fileName_ModelParams, and fileName_NullModelParams.

Usage

bitseq_fitGPs(gpData, fileName_logBF = NULL, fileName_ModelParams = NULL,
  fileName_NullModelParams = NULL)

Arguments

gpData	for example, output from bitseq_rnaSeqData.
fileName_logBF	name of the file which contains log Bayes factors.
fileName_ModelParams	name of the file which contains model parameters.
fileName_NullModelParams	name of the file which contains null model parameters.

Value

List of results

Author(s)

Hande Topa, [e-mail]

See Also

bitseq_rnaSeqData

Examples

## Not run: 
RNAseqDATA
gpData=RNAseqDATA$reltr
GPfits=bitseq_fitGPs(gpData)

## End(Not run)

GPrank documentation built on May 2, 2019, 3:35 p.m.