R/AllMethods.R
In RGLab/pepBayes: Bayesian modeling of peptide microarray assay responses
#' @title MCMC sampling for pepBayes models.
#'
#' @description Sample the posterior distribution of pepBayes models. The generic
#' function has methods for \link{peptideSet} objects and for \link{pepBayesFit}
#' objects for restarting a Markov chain from its last position.
#'
#' @details PepBayes estimates the posterior probability of
#' a differential antibody response against each peptide in a peptide microarray
#' assay, for each subject, when compared with control samples. Two common experimental
#' designs are automatically
#' detected. In an unpaired design, differences between a population of interest and
#' a control population are detected. A paired design has matched samples for each subject
#' before and after a treatment is applied. The function \code{pepBayesMcmc}
#' automatically detects the experimental design from the data, and
#' runs MCMC to sample the posterior distribution of the appropriate pepBayes model.
#'
#' When \code{x} is a \code{peptideSet} object, the paired or unpaired model is
#' chosen depending on the 'ptid' column of \code{pData(x)}. Paired data is detected
#' when each unique value of 'ptid' appears exactly twice among all slides.
#'
#' The \code{position_data} argument supplies information about peptide position
#' information. This argument may be a 'data.frame' with columns 'start',
#' 'end' or 'width', and 'peptide'. If a 'GRanges' object, then it must have either peptide
#' as a name or have peptide as a metadata column. If omitted (\code{position_data = NULL}),
#' response probability hyperparamters will be fixed during estimation.
#'
#' @param x A pepBayesFit or peptideSet object containing experiment data.
#' @param ... Additional arguments for sampling control. See methods.
#' @return An object of type \link{pepBayesFit} containing sorted slide metadata,
#' Markov chain posterior samples, posterior probabilities of binding, and
#' chain information.
#'
#' @seealso
#' \code{\link{pepBayesFit}}
#' \code{\link{pepBayesFit-methods}}
#' \code{\link{peptideSet}}
#' \code{\link{pepBayesEcm}}
#' \code{\link{pepBayesCalls}}
#'
#' @docType methods
#' @rdname pepBayesMcmc-methods
#' @exportMethod pepBayesMcmc
#' @useDynLib pepBayes
setGeneric("pepBayesMcmc", function(x, ...) standardGeneric("pepBayesMcmc"))
#' @rdname pepBayesMcmc-methods
#' @param position_data A data.frame or GRanges object with peptide position
#' information.
#' @param control_id Character, indicating which slides are control
#' slides from the "visit" column of the phenoData slot of the peptideSet argument.
#' @param n_samples Number of MCMC posterior samples to collect.
#' @param n_thin Number of MCMC iterations between posterior samples.
#' @param n_burn Number of MCMC iterations to perform before sampling begins.
#' @param n_threads Number of parallel threads to use during sampling.
#' @param prior_control_list A named list of parameter values to initialize
#' @param chain_control_list A named list of parameters controlling
#' which parameters to update or fix.
#' @aliases pepBayesMcmc,peptideSet-method
setMethod("pepBayesMcmc", signature = "peptideSet",
function(x, position_data = NULL, control_id = NULL,
n_samples, n_thin, n_burn, n_threads = detectCores(),
prior_control_list = NULL, chain_control_list = NULL) {
data_list = .joinPeptideData(x, position_data, control_id)
par_list = .initializeParameters(data_list, prior_control_list)
chain_list = .initializeChainList(n_samples, n_thin, n_burn,
data_list$method, n_threads, seed = 123, chain_control_list)
if (is.null(position_data))
chain_list$update_omega_hypers = FALSE
message(paste("Detected", data_list$method, "data"))
result = .Call("pepBayesMcmc", par_list, data_list,
chain_list, PACKAGE = "pepBayes")
result = .processMcmcOutput(result, data_list,
par_list, chain_list)
return(result)
})
#' @details When \code{x} is a pepBayesFit object, the chain is started from the most
#' recent parameter values in the fitted object. In the event that \code{x} is
#' the output of a previous MCMC fit, the argument \code{join_output} becomes
#' relevant. If \code{TRUE}, parameter traces are appended to those in \code{x}
#' and the posterior probabilities of binding are averaged together based on
#' the number of iterations in the fitted object and the current chain.
#' @rdname pepBayesMcmc-methods
#' @param join_output A boolean. Should input be merged with output?
#' @aliases pepBayesMcmc,pepBayesFit-method
setMethod("pepBayesMcmc", signature = "pepBayesFit",
function(x, n_samples, n_thin, n_burn = 0, n_threads = detectCores(),
join_output = TRUE) {
if (x@fitType == "ecm") {
# ecm algorithm is on normal, not logit scale.
om = x@algState$omega
om = ifelse(om < 1.0 && om > 0.0, log(om / 1 - om),
log((.001 + .99 * om) / (.999 - .99 * om)))
x@algState$omega = om
}
chain_list = .initializeChainList(n_samples, n_thin,
n_burn, x@slideInfo$method, n_threads, seed = 123, x@algInfo)
par_list = .initializeParameters(x@slideInfo, x@algState)
result = .Call("pepBayesMcmc", par_list, x@slideInfo,
chain_list, PACKAGE = "pepBayes")
result = .processMcmcOutput(result, x@slideInfo, x@algState,
chain_list)
if (join_output && x@fitType == "mcmc")
result = .joinMcmcOutput(x, result)
return(result)
})
#' @title Posterior maximization for pepBayes models.
#'
#' @description Maximize the posterior distribution of pepBayes models. The generic
#' function has methods for \link{peptideSet} objects and for \link{pepBayesFit}
#' objects for restarting an optimization from its last position.
#'
#' @details PepBayes estimates the posterior probability of
#' a differential antibody response against each peptide in a peptide microarray
#' assay, for each subject,
#' when compared with control samples. Two common experimental designs are automatically
#' detected. In an unpaired design, differences between a population of interest and
#' a control population are detected. A paired design has matched samples for each subject
#' before and after a treatment is applied. The function \code{pepBayesEcm}
#' automatically detects the experimental design from the data.
#'
#' When \code{x} is a \code{peptideSet} object, the paired or unpaired model is
#' chosen depending on the 'ptid' column of \code{pData(x)}. Paired data is detected
#' when each unique value of 'ptid' appears exactly twice among all slides.
#'
#' The \code{position_data} argument supplies information about peptide position
#' information. This argument may be a 'data.frame' with columns 'start',
#' 'end' or 'width', and 'peptide'. If a 'GRanges' object, then it must have either peptide
#' as a name or have peptide as a metadata column. Response probability hyperparamters are
#' fixed for EM estimation, but if position data are omitted (\code{position_data = NULL}),
#' they will also be fixed if the resulting fitted object is used to initialize a
#' \code{pepBayesMcmc}.
#'
#' @param x A pepBayesFit or peptideSet object containing experiment data.
#' @param ... Additional arguments for sampling control.
#' @return An object of type \link{pepBayesFit} containing sorted slide metadata,
#' parameter trace, posterior probabilities of binding, and
#' iteration information.
#'
#' @seealso
#' \code{\link{pepBayesFit}}
#' \code{\link{pepBayesFit-methods}}
#' \code{\link{peptideSet}}
#' \code{\link{pepBayesMcmc}}
#' \code{\link{pepBayesCalls}}
#'
#' @docType methods
#' @rdname pepBayesEcm-methods
#' @exportMethod pepBayesEcm
#' @useDynLib pepBayes
setGeneric("pepBayesEcm", function(x, ...) standardGeneric("pepBayesEcm"))
#' @param position_data A data.frame or GRanges object with peptide position
#' information.
#' @param n_iter The number of EM iterations to perform.
#' @param n_threads The number of parallel threads to use in the E-step.
#' @param n_rule The number of nodes to use for Gauss-Hermite numerical integration.
#' @param n_isamp The number of importance samples to estimate the E-step.
#' @param prior_control_list A named list controlling initial parameter values.
#' @param iter_control_list A named list controlling which parameters are updated.
#' @param schedule_iterations A boolean. Should the number of importance samples
#' gradually increase to \code{n_isamp} as iterations progress?
#' @rdname pepBayesEcm-methods
#' @aliases pepBayesEcm,peptideSet-method
setMethod("pepBayesEcm", signature = "peptideSet",
function(x, position_data = NULL, control_id = NULL,
n_iter = 30, n_threads = 1, n_rule = 10, n_isamp = 20,
prior_control_list = NULL, iter_control_list = NULL,
schedule_iterations = TRUE) {
data_list = .joinPeptideData(x, position_data, control_id,
n_rule)
par_list = .initializeParameters(data_list, prior_control_list)
iter_list = .initializeIterList(n_iter, n_threads, n_isamp,
iter_control_list)
iter_list$schedule_iterations = schedule_iterations
if (is.null(position_data))
iter_list$update_omega_hypers = FALSE
message(paste("Detected", data_list$method, "data"))
result = .Call("pepBayesEcm", par_list, data_list, iter_list,
PACKAGE = "pepBayes")
result = .processEcmOutput(result, data_list, par_list,
iter_list)
return(result)
})
#' @details When \code{x} is a pepBayesFit object, the optimization is started from the most
#' recent parameter values in the fitted object. In the event that \code{x} is
#' the output of a previous ECM fit, the argument \code{join_output} becomes
#' relevant. If \code{TRUE}, parameter traces are appended to those in \code{x}.
#' @param join_output A boolean. Should input be merged with output?
#' @rdname pepBayesEcm-methods
#' @aliases pepBayesEcm,pepBayesFit-method
setMethod("pepBayesEcm", signature = "pepBayesFit",
function(x, n_iter, n_threads = 1, n_rule = 10, n_isamp = 20,
schedule_iterations = FALSE, join_output = TRUE) {
if (x@fitType == "mcmc") {
# handle transformation of omega between algorithms.
om = 1 / (1 + exp(-x@algState$omega))
x@algState$omega = om
}
iter_list = .initializeIterList(n_iter, n_threads, n_isamp,
x@algInfo)
iter_list$schedule_iterations = schedule_iterations
par_list = .initializeParameters(x@slideInfo, x@algState)
result = .Call("pepBayesEcm", par_list, x@slideInfo,
iter_list, PACKAGE = "pepBayes")
result = .processEcmOutput(result, x@slideInfo, x@algState,
iter_list)
if (join_output && x@fitType == "ecm") {
result = .joinEcmOutput(x, result)
}
return(result)
})
#' @title pepBayesFit methods.
#'
#' @description Methods for interacting with \code{\link{pepBayesFit}} objects.
#'
#' @section Accessors:
#' \describe{
#' \item{\code{position(x)}}{Return the position of peptides.}
#' \item{\code{peptide(x)}}{Return the peptide amino acid sequences.}
#' \item{\code{parTrace(x)}}{Return trace of fitted parameter values across iterations
#' of fitting algorithm.}
#' \item{\code{ppb(x)}}{Return the estimated posterior probability of binding for each subject
#' /peptide combination.}
#' }
#'
#' @section pepBayes Analysis:
#' \describe{
#' \item{\code{pepBayesCalls(x, fdr)}}{Return calls for a pepBayesFit at the desired
#' \code{fdr} value.}
#' \item{\code{pepBayesMcmc(x, ...)}}{Run \code{\link{pepBayesMcmc}}, starting the chain
#' at the last position of the parameters in \code{x}.}
#' \item{\code{pepBayesEcm(x, ...)}}{Run \code{\link{pepBayesEcm}}, starting the optimization
#' at the last position of the parameters in \code{x}.}
#' }
#'
#' @seealso
#' \code{\link{pepBayesEcm}}
#' \code{\link{pepBayesMcmc}}
#' \code{\link{pepBayesCalls}}
#'
#' @name pepBayesFit-methods
#' @rdname pepBayesFit-methods
#' @exportMethod "position"
#' @exportMethod "peptide"
#' @exportMethod "parTrace"
#' @exportMethod "ppb"
NULL
setMethod("position", signature = "pepBayesFit",
function(x) {
return(x@slideInfo$orig_pos)
})
setMethod("peptide", signature = "pepBayesFit",
function(x) {
return(x@slideInfo$peptide)
})
setGeneric("parTrace", function(x, ...) standardGeneric("parTrace"))
setMethod("parTrace", signature = "pepBayesFit",
function(x) {
return(x@parameterOutput)
})
setGeneric("ppb", function(x, ...) standardGeneric("ppb"))
setMethod("ppb", signature = "pepBayesFit",
function(x) {
return(x@ppb)
})
#' @title Call peptide binding from pepBayes output.
#'
#' @description Use posterior probability of response to generate a
#' call threshold that controls the false discovery rate of pepBayes positivity
#' calls.
#'
#' @details Probabilities are sorted in descending order. The largest integer K
#' is selected such that the mean of the first K sorted probability values exceeds
#' the value (1 - \code{fdr}). The probability threshold is chosen to be the K-th
#' sorted probability value. Entries with probability greater than or equal to the
#' threshold value are considered positive calls, else an entry is a negative call.
#'
#' @param x A matrix or pepBayesFit object.
#' @param fdr Desired false discovery rate.
#'
#' @return A matrix of booleans with TRUE denoting a positive call, and FALSE
#' a negative call.
#'
#' @rdname pepBayesCalls-methods
#' @docType methods
#' @exportMethod pepBayesCalls
#' @seealso
#' \code{\link{pepBayesEcm}}
#' \code{\link{pepBayesMcmc}}
#' \code{\link{pepBayesFit-methods}}
setGeneric("pepBayesCalls", function(x, fdr, ...) standardGeneric("pepBayesCalls"))
#' @rdname pepBayesCalls-methods
#' @aliases pepBayesCalls,pepBayesFit,numeric-method
setMethod("pepBayesCalls", signature = c("pepBayesFit", "numeric"),
function(x, fdr = .05) {
return(.getCalls(ppb(x), fdr))
})
#' @rdname pepBayesCalls-methods
#' @aliases pepBayesCalls,matrix,numeric-method
setMethod("pepBayesCalls", signature = c("matrix", "numeric"),
function(x, fdr = .05) {
return(.getCalls(x, fdr))
})
RGLab/pepBayes documentation built on May 8, 2019, 5:55 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
#' @title MCMC sampling for pepBayes models.
#'
#' @description Sample the posterior distribution of pepBayes models. The generic
#' function has methods for \link{peptideSet} objects and for \link{pepBayesFit}
#' objects for restarting a Markov chain from its last position.
#'
#' @details PepBayes estimates the posterior probability of
#' a differential antibody response against each peptide in a peptide microarray
#' assay, for each subject, when compared with control samples. Two common experimental
#' designs are automatically
#' detected. In an unpaired design, differences between a population of interest and
#' a control population are detected. A paired design has matched samples for each subject
#' before and after a treatment is applied. The function \code{pepBayesMcmc}
#' automatically detects the experimental design from the data, and
#' runs MCMC to sample the posterior distribution of the appropriate pepBayes model.
#'
#' When \code{x} is a \code{peptideSet} object, the paired or unpaired model is
#' chosen depending on the 'ptid' column of \code{pData(x)}. Paired data is detected
#' when each unique value of 'ptid' appears exactly twice among all slides.
#'
#' The \code{position_data} argument supplies information about peptide position
#' information. This argument may be a 'data.frame' with columns 'start',
#' 'end' or 'width', and 'peptide'. If a 'GRanges' object, then it must have either peptide
#' as a name or have peptide as a metadata column. If omitted (\code{position_data = NULL}),
#' response probability hyperparamters will be fixed during estimation.
#'
#' @param x A pepBayesFit or peptideSet object containing experiment data.
#' @param ... Additional arguments for sampling control. See methods.
#' @return An object of type \link{pepBayesFit} containing sorted slide metadata,
#' Markov chain posterior samples, posterior probabilities of binding, and
#' chain information.
#'
#' @seealso
#' \code{\link{pepBayesFit}}
#' \code{\link{pepBayesFit-methods}}
#' \code{\link{peptideSet}}
#' \code{\link{pepBayesEcm}}
#' \code{\link{pepBayesCalls}}
#'
#' @docType methods
#' @rdname pepBayesMcmc-methods
#' @exportMethod pepBayesMcmc
#' @useDynLib pepBayes
setGeneric("pepBayesMcmc", function(x, ...) standardGeneric("pepBayesMcmc"))
#' @rdname pepBayesMcmc-methods
#' @param position_data A data.frame or GRanges object with peptide position
#' information.
#' @param control_id Character, indicating which slides are control
#' slides from the "visit" column of the phenoData slot of the peptideSet argument.
#' @param n_samples Number of MCMC posterior samples to collect.
#' @param n_thin Number of MCMC iterations between posterior samples.
#' @param n_burn Number of MCMC iterations to perform before sampling begins.
#' @param n_threads Number of parallel threads to use during sampling.
#' @param prior_control_list A named list of parameter values to initialize
#' @param chain_control_list A named list of parameters controlling
#' which parameters to update or fix.
#' @aliases pepBayesMcmc,peptideSet-method
setMethod("pepBayesMcmc", signature = "peptideSet",
function(x, position_data = NULL, control_id = NULL,
n_samples, n_thin, n_burn, n_threads = detectCores(),
prior_control_list = NULL, chain_control_list = NULL) {
data_list = .joinPeptideData(x, position_data, control_id)
par_list = .initializeParameters(data_list, prior_control_list)
chain_list = .initializeChainList(n_samples, n_thin, n_burn,
data_list$method, n_threads, seed = 123, chain_control_list)
if (is.null(position_data))
chain_list$update_omega_hypers = FALSE
message(paste("Detected", data_list$method, "data"))
result = .Call("pepBayesMcmc", par_list, data_list,
chain_list, PACKAGE = "pepBayes")
result = .processMcmcOutput(result, data_list,
par_list, chain_list)
return(result)
})
#' @details When \code{x} is a pepBayesFit object, the chain is started from the most
#' recent parameter values in the fitted object. In the event that \code{x} is
#' the output of a previous MCMC fit, the argument \code{join_output} becomes
#' relevant. If \code{TRUE}, parameter traces are appended to those in \code{x}
#' and the posterior probabilities of binding are averaged together based on
#' the number of iterations in the fitted object and the current chain.
#' @rdname pepBayesMcmc-methods
#' @param join_output A boolean. Should input be merged with output?
#' @aliases pepBayesMcmc,pepBayesFit-method
setMethod("pepBayesMcmc", signature = "pepBayesFit",
function(x, n_samples, n_thin, n_burn = 0, n_threads = detectCores(),
join_output = TRUE) {
if (x@fitType == "ecm") {
# ecm algorithm is on normal, not logit scale.
om = x@algState$omega
om = ifelse(om < 1.0 && om > 0.0, log(om / 1 - om),
log((.001 + .99 * om) / (.999 - .99 * om)))
x@algState$omega = om
}
chain_list = .initializeChainList(n_samples, n_thin,
n_burn, x@slideInfo$method, n_threads, seed = 123, x@algInfo)
par_list = .initializeParameters(x@slideInfo, x@algState)
result = .Call("pepBayesMcmc", par_list, x@slideInfo,
chain_list, PACKAGE = "pepBayes")
result = .processMcmcOutput(result, x@slideInfo, x@algState,
chain_list)
if (join_output && x@fitType == "mcmc")
result = .joinMcmcOutput(x, result)
return(result)
})
#' @title Posterior maximization for pepBayes models.
#'
#' @description Maximize the posterior distribution of pepBayes models. The generic
#' function has methods for \link{peptideSet} objects and for \link{pepBayesFit}
#' objects for restarting an optimization from its last position.
#'
#' @details PepBayes estimates the posterior probability of
#' a differential antibody response against each peptide in a peptide microarray
#' assay, for each subject,
#' when compared with control samples. Two common experimental designs are automatically
#' detected. In an unpaired design, differences between a population of interest and
#' a control population are detected. A paired design has matched samples for each subject
#' before and after a treatment is applied. The function \code{pepBayesEcm}
#' automatically detects the experimental design from the data.
#'
#' When \code{x} is a \code{peptideSet} object, the paired or unpaired model is
#' chosen depending on the 'ptid' column of \code{pData(x)}. Paired data is detected
#' when each unique value of 'ptid' appears exactly twice among all slides.
#'
#' The \code{position_data} argument supplies information about peptide position
#' information. This argument may be a 'data.frame' with columns 'start',
#' 'end' or 'width', and 'peptide'. If a 'GRanges' object, then it must have either peptide
#' as a name or have peptide as a metadata column. Response probability hyperparamters are
#' fixed for EM estimation, but if position data are omitted (\code{position_data = NULL}),
#' they will also be fixed if the resulting fitted object is used to initialize a
#' \code{pepBayesMcmc}.
#'
#' @param x A pepBayesFit or peptideSet object containing experiment data.
#' @param ... Additional arguments for sampling control.
#' @return An object of type \link{pepBayesFit} containing sorted slide metadata,
#' parameter trace, posterior probabilities of binding, and
#' iteration information.
#'
#' @seealso
#' \code{\link{pepBayesFit}}
#' \code{\link{pepBayesFit-methods}}
#' \code{\link{peptideSet}}
#' \code{\link{pepBayesMcmc}}
#' \code{\link{pepBayesCalls}}
#'
#' @docType methods
#' @rdname pepBayesEcm-methods
#' @exportMethod pepBayesEcm
#' @useDynLib pepBayes
setGeneric("pepBayesEcm", function(x, ...) standardGeneric("pepBayesEcm"))
#' @param position_data A data.frame or GRanges object with peptide position
#' information.
#' @param n_iter The number of EM iterations to perform.
#' @param n_threads The number of parallel threads to use in the E-step.
#' @param n_rule The number of nodes to use for Gauss-Hermite numerical integration.
#' @param n_isamp The number of importance samples to estimate the E-step.
#' @param prior_control_list A named list controlling initial parameter values.
#' @param iter_control_list A named list controlling which parameters are updated.
#' @param schedule_iterations A boolean. Should the number of importance samples
#' gradually increase to \code{n_isamp} as iterations progress?
#' @rdname pepBayesEcm-methods
#' @aliases pepBayesEcm,peptideSet-method
setMethod("pepBayesEcm", signature = "peptideSet",
function(x, position_data = NULL, control_id = NULL,
n_iter = 30, n_threads = 1, n_rule = 10, n_isamp = 20,
prior_control_list = NULL, iter_control_list = NULL,
schedule_iterations = TRUE) {
data_list = .joinPeptideData(x, position_data, control_id,
n_rule)
par_list = .initializeParameters(data_list, prior_control_list)
iter_list = .initializeIterList(n_iter, n_threads, n_isamp,
iter_control_list)
iter_list$schedule_iterations = schedule_iterations
if (is.null(position_data))
iter_list$update_omega_hypers = FALSE
message(paste("Detected", data_list$method, "data"))
result = .Call("pepBayesEcm", par_list, data_list, iter_list,
PACKAGE = "pepBayes")
result = .processEcmOutput(result, data_list, par_list,
iter_list)
return(result)
})
#' @details When \code{x} is a pepBayesFit object, the optimization is started from the most
#' recent parameter values in the fitted object. In the event that \code{x} is
#' the output of a previous ECM fit, the argument \code{join_output} becomes
#' relevant. If \code{TRUE}, parameter traces are appended to those in \code{x}.
#' @param join_output A boolean. Should input be merged with output?
#' @rdname pepBayesEcm-methods
#' @aliases pepBayesEcm,pepBayesFit-method
setMethod("pepBayesEcm", signature = "pepBayesFit",
function(x, n_iter, n_threads = 1, n_rule = 10, n_isamp = 20,
schedule_iterations = FALSE, join_output = TRUE) {
if (x@fitType == "mcmc") {
# handle transformation of omega between algorithms.
om = 1 / (1 + exp(-x@algState$omega))
x@algState$omega = om
}
iter_list = .initializeIterList(n_iter, n_threads, n_isamp,
x@algInfo)
iter_list$schedule_iterations = schedule_iterations
par_list = .initializeParameters(x@slideInfo, x@algState)
result = .Call("pepBayesEcm", par_list, x@slideInfo,
iter_list, PACKAGE = "pepBayes")
result = .processEcmOutput(result, x@slideInfo, x@algState,
iter_list)
if (join_output && x@fitType == "ecm") {
result = .joinEcmOutput(x, result)
}
return(result)
})
#' @title pepBayesFit methods.
#'
#' @description Methods for interacting with \code{\link{pepBayesFit}} objects.
#'
#' @section Accessors:
#' \describe{
#' \item{\code{position(x)}}{Return the position of peptides.}
#' \item{\code{peptide(x)}}{Return the peptide amino acid sequences.}
#' \item{\code{parTrace(x)}}{Return trace of fitted parameter values across iterations
#' of fitting algorithm.}
#' \item{\code{ppb(x)}}{Return the estimated posterior probability of binding for each subject
#' /peptide combination.}
#' }
#'
#' @section pepBayes Analysis:
#' \describe{
#' \item{\code{pepBayesCalls(x, fdr)}}{Return calls for a pepBayesFit at the desired
#' \code{fdr} value.}
#' \item{\code{pepBayesMcmc(x, ...)}}{Run \code{\link{pepBayesMcmc}}, starting the chain
#' at the last position of the parameters in \code{x}.}
#' \item{\code{pepBayesEcm(x, ...)}}{Run \code{\link{pepBayesEcm}}, starting the optimization
#' at the last position of the parameters in \code{x}.}
#' }
#'
#' @seealso
#' \code{\link{pepBayesEcm}}
#' \code{\link{pepBayesMcmc}}
#' \code{\link{pepBayesCalls}}
#'
#' @name pepBayesFit-methods
#' @rdname pepBayesFit-methods
#' @exportMethod "position"
#' @exportMethod "peptide"
#' @exportMethod "parTrace"
#' @exportMethod "ppb"
NULL
setMethod("position", signature = "pepBayesFit",
function(x) {
return(x@slideInfo$orig_pos)
})
setMethod("peptide", signature = "pepBayesFit",
function(x) {
return(x@slideInfo$peptide)
})
setGeneric("parTrace", function(x, ...) standardGeneric("parTrace"))
setMethod("parTrace", signature = "pepBayesFit",
function(x) {
return(x@parameterOutput)
})
setGeneric("ppb", function(x, ...) standardGeneric("ppb"))
setMethod("ppb", signature = "pepBayesFit",
function(x) {
return(x@ppb)
})
#' @title Call peptide binding from pepBayes output.
#'
#' @description Use posterior probability of response to generate a
#' call threshold that controls the false discovery rate of pepBayes positivity
#' calls.
#'
#' @details Probabilities are sorted in descending order. The largest integer K
#' is selected such that the mean of the first K sorted probability values exceeds
#' the value (1 - \code{fdr}). The probability threshold is chosen to be the K-th
#' sorted probability value. Entries with probability greater than or equal to the
#' threshold value are considered positive calls, else an entry is a negative call.
#'
#' @param x A matrix or pepBayesFit object.
#' @param fdr Desired false discovery rate.
#'
#' @return A matrix of booleans with TRUE denoting a positive call, and FALSE
#' a negative call.
#'
#' @rdname pepBayesCalls-methods
#' @docType methods
#' @exportMethod pepBayesCalls
#' @seealso
#' \code{\link{pepBayesEcm}}
#' \code{\link{pepBayesMcmc}}
#' \code{\link{pepBayesFit-methods}}
setGeneric("pepBayesCalls", function(x, fdr, ...) standardGeneric("pepBayesCalls"))
#' @rdname pepBayesCalls-methods
#' @aliases pepBayesCalls,pepBayesFit,numeric-method
setMethod("pepBayesCalls", signature = c("pepBayesFit", "numeric"),
function(x, fdr = .05) {
return(.getCalls(ppb(x), fdr))
})
#' @rdname pepBayesCalls-methods
#' @aliases pepBayesCalls,matrix,numeric-method
setMethod("pepBayesCalls", signature = c("matrix", "numeric"),
function(x, fdr = .05) {
return(.getCalls(x, fdr))
})
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.