poe.mcmc: Probability of Expression (POE)

Description Usage Arguments Value Author(s) References Examples

Description

Differential expression using latent categories of up/down regulation. Three component normal-uniform mixture model under Bayesian hierarchical analysis. This is a C implementation of poe.fit function of POE package (MCMC). R portion of source code was directly adapted from POE; poe.one.iteration function was re-written in C. Some of the optional arguments available in poe.fit are suppressed here, and you cannot save the chain of samples drawn for numerical integration.

Usage

1
2
3
4
5
6
7
8
9
  poe.mcmc(AA, NN = NULL, id = NULL, M = 2000, kap.min=3.0,
      logdata=FALSE, stepsize=0.5, 
      centersample = FALSE, centergene = FALSE, generatestarts = TRUE, start.method = 1,
      startobject = R0, collapse.to.two = FALSE, burnin=200,
      collapse.window=50, converge.threshold=0.01,
      PR = list(alpha.mm = 0, alpha.sd = 100, mu.mm = 0, mu.sd = 100,
          pipos.mm = 0, pipos.sd = 100, pineg.mm = 0, pineg.sd = 100,
          kap.pri.rate = 1, tausqinv.aa = 1, tausqinv.bb = 0.1))
  # poe.mcmc(AA = chen, NN = 1 - chen.spl$metastasis, M=2000)

Arguments

AA

Matrix or exprs from an ExpressionSet object.

NN

Phenotypic label of arrays. Samples from the reference group (normal phenotype) must be labeled as 1. For instance, if you wish to identify differentially expressed genes between cancer and normal samples, mark normal as 1 and cancer as 0. Then the high probability hits indicate over or under-expressed genes in specific samples with respect to the gene expression distribution of the normal samples. Alternatively, you may leave this blank if you wish to get the probability of high and low expression from the mixture distribution using all samples not relying on the known sample information.

M

Number of MCMC iterations after burn-in.

Value

poe

Probability of over/under expression. Transformed gene expression on a fixed scale of [-1,1].

Other values

Posterior median estimates of parameters. Please refer to POE package for details.

Author(s)

Debashis Ghosh <[email protected]>, Hyungwon Choi <[email protected]>

References

G. Parmigiani et al, JRSS, 64:717-736, 2002 or URL: http://astor.som.jhmi.edu/poe/

Examples

1
2
  # poe.mat <- poe.mcmc(AA=exprmat, NN=clvec, M=10000)
  # One can also provide different hyperparameter values. 

Example output



metaArray documentation built on Nov. 1, 2018, 2:27 a.m.