bandle-internal: sample allocations, probabilities and compute loglikilihoods

In ococrook/bandle: An R package for the Bayesian analysis of differential subcellular localisation experiments

sample allocations, probabilities and compute loglikilihoods

Description

Internal sampling function, not for outside use documented for completness

Usage

proteinAllocation(loglikelihoods, currentweights, alloctemp, cond)

outlierAllocationProbs(
  outlierlikelihood,
  loglikelihoods,
  epsilon,
  alloctemp,
  cond
)

sampleOutlier(allocoutlierprob)

covOrganelle(object, fcol = "markers")

pg_prior(object_cond1, object_cond2, K, pgPrior = NULL, fcol = "markers")

sample_weights_pg(nk_mat, pgPrior, w, K, tau = 0.2)

sample_weights_dir(nk_mat, dirPrior)

Arguments

loglikelihoods	the log likelihoods
currentweights	the current allocations weights
alloctemp	the current protein allocations
cond	the control = 1, treatment = 2
outlierlikelihood	the outlier log likelihoods
epsilon	the outlier component weight
allocoutlierprob	the outlier probabilities
object	An instance of class MSnSet
fcol	The feature column containing the markers.
object_cond1	A list of instance of class MSnSets usually control
object_cond2	A list of instance of class MSnSets usually treatment
K	The number of organelle classes
pgPrior	The Polya-Gamma prior
nk_mat	The summary matrix of allocations
w	The Polya-Gamma auxiliary variable
tau	The empirical bayes parameter for the Polya-Gamma variable. Defaults to 0.2.
dirPrior	The Dirichlet prior

Value

returns samples for protein allocations, log likelihoods and probabilities

returns outlier probabilities

returns outlier allocations

returns covariance of organelles using marker proteins

returns the Polya-Gamma prior

returns A sample of the weights using Polya-Gamma priors.

returns A sample of the weights using Dirichlet prior.

Examples

library(pRolocdata)
data("tan2009r1")
covOrganelle(object = tan2009r1)


library(pRolocdata)
data("tan2009r1")
set.seed(1)
tansim <- sim_dynamic(object = tan2009r1, 
                    numRep = 6L,
                   numDyn = 100L)
d1 <- tansim$lopitrep
control1 <- d1[1:3]
treatment1 <- d1[4:6]
out <- pg_prior(object_cond1 = control1,
 object_cond2 = treatment1, K = 11) 

ococrook/bandle documentation built on Nov. 4, 2024, 12:27 a.m.