processFCS: Import .fcs file data into a 'SingleCellExperiment' object...
In kordastilab/ImmunoCluster: Fast and accesible Immune profiling CyTOF data

Description Usage Arguments Value Author(s) Examples

Import .fcs file data into a SingleCellExperiment object with experiment metadata.

processFCS(
  files,
  assayname = "scaled",
  metadata = NULL,
  filter = FALSE,
  bgNoiseThreshold = 1,
  euclideanNormThreshold = 1,
  transformation = TRUE,
  transFun = function(x) asinh(x),
  asinhFactor = 5,
  downsample = 1e+05,
  downsample_grouping = "group",
  downsampleVar = NULL,
  colsDiscard = c("Time", "Event_length", "Center", "Offset", "Width", "Residual",
    "tSNE1", "tSNE2", "BCKG"),
  colsRetain = NULL,
  newColnames = NULL
)

`files`	a character string representing the filenames of the .fcs files to import.
`assayname`	a character string to name the assay within the `SingleCellExperiment`.
`metadata`	the metadata must be a dataframe where each row represents an individial .fcs file to be incorporated into the SCE object. The number of rows must match the number of files. Each column represents a metadata attribute stored in the SCE.
`filter`	a boolean of whether to perform noise filtering and normalization on the data.
`bgNoiseThreshold`	an numeric representing the background noise threshold to filter.
`euclideanNormThreshold`	an numeric as input to euclidean normalization function.
`transformation`	a boolean of whether to transform the data.
`asinhFactor`	a numeric that will be used as the cofactor in asinh transformation.
`downsample`	a numeric defining the number of cells to downsample to from all cells (if downsample_grouping is NULL).
`downsample_grouping`	a character string specifying a metadata slot from which to downsample evenly by the numeric specified in downsample. n downsampled cells will be chosen evenly from each downsample_grouping.
`downsampleVar`	a boolean defining whether to donwsample from all cells by variance. Does not work in combination with downsample_grouping.
`colsDiscard`	a vector of character strings of column names to discard.
`colsRetain`	a vector of character strings of column names to retain, inverse of colsDiscard.
`newColnames`	a vector of character strings to rename the channel parameters from the raw .fcs data. The order of newColnames must match the the exact channels to be renamed in order after filtering. If colsRetain or colsDiscard has been used to filter out certain channels this needs to be taken into account when renaming the subsequent filtered channel list or the new names may be attached to the wrong channels.
`TransFun`	a funtion to perform transformation on the data with, is by default the asinh with cofactor 5.

a SingleCellExperiment object.

Kevin Blighe, James Opzoomer james.opzoomer@kcl.ac.uk

# Create SCE object with CyTOF data inside
require(SingleCellExperiment)
sce_gvhd = processFCS(
 files = filelist,
 metadata = metadata,
 filter = FALSE, # Do not perform noise filtering
 transformation = TRUE, # Transform data
 transFun = function (x) asinh(x), # Using asinh function (cofactor 5 is used by default)
 downsampleVar = NULL, # No downsampling by variance
 downsample = NULL, # Downsample to n cells per downsample_grouping
 downsample_grouping = NULL, # if downsample = n, downsample n cells by the specified metadata grouping.
 newColnames = col_names, # Rename columns from panel metadata
 colsDiscard = colsDiscard) # Discard columns not selected for downstream analysis