R/NetworkProbExp.R

In EDISON: Network Reconstruction and Changepoint Detection

#' Calculates the prior probability of the network using the exponential prior.
#' 
#' This function calculates the log prior probability of the network structure.
#' It uses the exponential information sharing prior.
#' 
#' 
#' @param network.info Network information collected using the function
#' \code{\link{CollectNetworkInfo}}
#' @return Returns the log prior probability of the network segments.
#' @author Frank Dondelinger
#' @seealso \code{\link{NetworkRatioExp}}, \code{\link{CalculatePriorRatio}}
#' @references For information about the exponential information sharing prior,
#' see:
#' 
#' Husmeier et al. (2010), "Inter-time segment information sharing for
#' non-homogeneous dynamic Bayesian networks", NIPS.
#' 
#' Dondelinger et al. (2012), "Non-homogeneous dynamic Bayesian networks with
#' Bayesian regularization for inferring gene regulatory networks with
#' gradually time-varying structure", Machine Learning.
#' @export NetworkProbExp
NetworkProbExp <-
function(network.info) {
  # Calculate the (log) probability P(M_i|M_{i-1})*P(M_{i+1}|M_i), where M_i is
  # network segment being changed.
  #
  # Args:
  #   net:          The network being changed
  #   network_info: The network structures and associated information.
  #             network.info$nets         - Structure of all segments 
  #             network.info$prior.params - Beta parameters for all segments
  #             network.info$segment      - Segment being changed
  #             network.info$target       - Target node whose edge is being changed
  #             network.info$parent       - Parent being changed
  # Returns:
  #   P(M_i|M_{i-1})*P(M_{i+1}|M_i)
  
  target = network.info$target

  suff.statistics = CalculateChanges(network.info, 'soft')
 
  differences = suff.statistics[2] + suff.statistics[3]
  
  beta = network.info$prior.params[target]
   
  logprior = -beta*differences
 
  return(logprior)
}