Generate a dynamic spatial proteomics experiment

A function to simulate dynamic spatial proteomics data using a bootstrap method

sim_dynamic(
  object,
  subsample = NULL,
  knn_par = 10L,
  fcol = "markers",
  numRep = 6L,
  method = "wild",
  batch = FALSE,
  frac_perm = FALSE,
  nu = 2,
  numDyn = 20L
)

Arguments

object: A instance of class MSnSet from which to generate a spatial proteomics dataset.
subsample: how many proteins to subsample to speed up analysis. Default is NULL.
knn_par: the number of nearest neighbours to use in KNN classification to simulate dataset. Default is 10
fcol: feature column to indicate markers. Default is "markers". Proteins with unknown localisations must be encoded as "unknown".
numRep: The total number of datasets to generate. Default is 6. An integer must be provided
method: The bootstrap method to use to simulate dataset. Default is "wild". refer to BANDLE paper for more details.
batch: Whether or not to include batch effects. Default is FALSE.
frac_perm: whether or not to permute the fractions. Default is FALSE
nu: parameter to generate residual inflated noise. Default is 2. See BANDLE paper for more details
numDyn: An integer number of protein to simulate dynamic transitions. Default is 20

Value

returns simulate dynamic lopit datasets and the name of the relocalated protein.

Examples

library(pRolocdata)
data("tan2009r1")
set.seed(1)
tansim <- sim_dynamic(object = tan2009r1, numRep = 6L, numDyn = 100L)