scHOT: scHOT
In scHOT: single-cell higher order testing

Description Usage Arguments Value Examples

A wrapper function to perform scHOT

scHOT(
  scHOT,
  testingScaffold = NULL,
  weightMatrix = NULL,
  positionType = NULL,
  positionColData = NULL,
  nrow.out = NULL,
  averageAcrossTrajectoryTies = FALSE,
  higherOrderFunction = NULL,
  higherOrderFunctionType = NULL,
  numberPermutations = 1000,
  numberScaffold = 100,
  storePermutations = TRUE,
  higherOrderSummaryFunction = sd,
  parallel = FALSE,
  BPPARAM = BiocParallel::SerialParam(),
  usenperm_estimate = FALSE,
  nperm_estimate = 10000,
  maxDist = 0.1,
  plot = FALSE,
  verbose = TRUE,
  ...
)

`scHOT`	A scHOT object
`testingScaffold`	A matrix with rows for each testing combination
`weightMatrix`	A matrix indicates the weight matrix for scHOT analysis
`positionType`	A string indicating the position type, either "trajectory" or "spatial"
`positionColData`	Either trajectory or spatial information for each sample. If positionType is "trajectory" then positionColData should be a character or numeric indicating the subset of colData of the scHOT object. If positionType is "spatial" then positionColData should be a character or numeric vector indicating the subset of colData that give the full spatial coordinates.
`nrow.out`	The number of weightings to include for testing, a smaller value is faster for computation
`averageAcrossTrajectoryTies`	Logical indicating whether ties in the trajectory should be given the same local weights
`higherOrderFunction`	A function object indicates the higher order function
`higherOrderFunctionType`	is "weighted" or "unweighted", determines if there is a weighting argument in the higher order function
`numberPermutations`	The number of permutations, set as 1000 by default
`numberScaffold`	The number of testing scaffolds to perform permutations, set as 100 by default
`storePermutations`	a logical flag on whether permutation values should be saved
`higherOrderSummaryFunction`	A functon indicating the higher order summary function (default is standard deviation 'sd')
`parallel`	A logical input indicating whether to run the permutation test using multiple cores in parallel.
`BPPARAM`	A `BiocParallelParam` class object from the `BiocParallel` package is used. Default is SerialParam().
`usenperm_estimate`	Logical (default FALSE) if number of neighbouring permutations should be used to estimate P-values, or if difference of global higher order statistic should be used
`nperm_estimate`	Number of neighbouring permutations to use for p-value estimation
`maxDist`	max difference of global higher order statistic to use for p-value estimation (default 0.1)
`plot`	A logical input indicating whether the results are plotted
`verbose`	A logical input indicating whether the intermediate steps will be printed
`...`	parameters for function trajectoryWeightMatrix or spatialWeightMatrix

A scHOT object

data(MOB_subset)
sce_MOB_subset <- MOB_subset$sce_MOB_subset
scHOT_spatial <- scHOT_buildFromSCE(sce_MOB_subset,
                                    assayName = "logcounts",
                                    positionType = "spatial",
                                    positionColData = c("x", "y"))
pairs <- matrix(c("Arrb1", "Mtor", "Dnm1l", "Gucy1b3"), ncol = 2, byrow = TRUE)
rownames(pairs) <- apply(pairs,1,paste0,collapse = "_")

scHOT_spatial <- scHOT(scHOT_spatial,
                       testingScaffold = pairs,
                       positionType = "spatial",
                       positionColData = c("x", "y"),
                       nrow.out = NULL,
                       higherOrderFunction = weightedSpearman,
                       higherOrderFunctionType = "weighted",
                       numberPermutations = 100,
                       higherOrderSummaryFunction = sd,
                       parallel = FALSE,
                       verbose = TRUE,
                       span = 0.05)