cytofin_normalize_nrs: Batch normalize CyTOF plates from heterogeneous sources using...
In bennyyclo/Cytofin: Integrate CyTOF Datasets From Heterogeneous Sources

This function batch normalizes CyTOF data from multiple plates (from one or more experimental cohorts) by computing the non-redundancy score (NRS) for each channel in the dataset, then using the most redundant (i.e. the "most stable") channels as a reference for batch normalization.

cytofin_normalize_nrs(
  metadata_path,
  panel_path,
  input_data_path,
  output_data_path,
  input_prefix = "homogenized_",
  output_prefix = "normalized_",
  shift_factor = 0,
  scale_factor = 0.2,
  nchannels = 3,
  make_plot = FALSE
)

`metadata_path`	A filepath leading to an .xlsx or .csv file containing a table of CyTOF file (.fcs file) names. Columns should include `filename`, `cohort`, `plate_number`, `patient_id`, `condition`, `is_anchor`, and `validation`. See the vignette for details: `vignette("help", package = "cytofin")`
`panel_path`	A file path leading to an .xlsx or .csv file containing a table of standardized antigen panel information. Columns should include `metal_name`, `antigen_name`, `antigen_pattern`, `lineage`, `functional`, and `general`. See the vignette for details: `vignette("help", package = "cytofin")`
`input_data_path`	A folder directory containing the input CyTOF files to be normalized. In most cases, this will be the directory to which the output .fcs files from `cytofin_homogenize` were written.
`output_data_path`	A folder directory to which the output (i.e. batch normalized/batch corrected) .fcs files should be written.
`input_prefix`	The string that was appended to the name of the input files of `cytofin_homogenize` to create their corresponding output file names. Defaults to "homogenized_".
`output_prefix`	A string to be appended to the name of each input file to create the name of the corresponding output file (post-homogenization). Defaults to "normalized_" (e.g. an input file named "file1.fcs" will correspond to the output file "normalized_file1.fcs" saved in `output_data_path`).
`shift_factor`	The scalar value `a` in the following equation used to transform CyTOF raw data ion counts using the hyperbolic arcsinh function: `new_x <- asinh(a + b * x)`. Defaults to 0.
`scale_factor`	The scalar value `b` in the following equation used to transform CyTOF raw data ion counts using the hyperbolic arcsinh function: `new_x <- asinh(a + b * x)`. Defaults to 0.2.
`nchannels`	An integer representing the number of most stable channels to use during batch normalization. Defaults to 3.
`make_plot`	A boolean value indicating if a plot depicting the non- redundancy scores of each marker in each .fcs file being batch normalized should be plotted as a side-effect of the function call. Defaults to FALSE.

Batch-normalized .fcs files are saved in the directory specified by output_data_path.

In addition, a data.frame containing information about each input .fcs file (that can be used for plotting with cytofin_make_plots) is returned with the following columns:

All of the columns in the input metadata table (located at metadata_path)
universal_mean: the universal mean vector to which all files are adjusted (will be identical for all input .fcs files)
universal_var: the universal mean vector to which all files are adjusted (will be identical for all input .fcs files)
anchor_mean: the mean (across all cells) vector for the anchor file associated with each input .fcs file (i.e. the anchor located on the same plate as the input .fcs file)
anchor_var: the variance (across all cells) vector for the anchor file associated with each input .fcs file (i.e. the anchor located on the same plate as the input .fcs file)
mean_b4norm: the mean (across all cells) vector of the input .fcs file before batch normalization.
var_b4norm: the variance (across all cells) vector of the input .fcs file before batch normalization.
mean_norm: the mean (across all cells) vector of the input .fcs file after batch normalization.
var_norm: the variance (across all cells) vector of the input .fcs file after batch normalization.
anchor_mean_norm: the mean (across all cells) vector for the anchor file associated with each input .fcs file (i.e. the anchor located on the same plate as the input .fcs file) after batch normalization.
anchor_var_norm: the variance (across all cells) vector for the anchor file associated with each input .fcs file (i.e. the anchor located on the same plate as the input .fcs file) after batch normalization.