tune_conumee_pipe: Run Conumee pipeline with different preprocessing methods.
In markgene/yamatCN: Yet Another Methylation Array Toolkit Copy Number Analysis

Description Usage Arguments Details Value

Run Conumee pipeline with different preprocessing methods, i.e. the combination of normalization methods and batch effect correction. The function serves as a template for writing your own function to run the pipeline with different parameters, rather than a function of general purpose.

tune_conumee_pipe(ref, qry, report_dir, norm_methods = c("swan",
  "illumina", "raw", "quantile", "noob", "funnorm", "yamat", "dkfz",
  "methylcnv"), batch = NULL, batch2 = NULL, pipe_file = NULL,
  dnacopy_seed = 1, segment_grid, overwrite = FALSE, verbose = TRUE,
  ...)

`ref`	Reference samples stored in an object of `RGChannelSet-class`.
`qry`	Query samples stored in an object of `RGChannelSet-class`.
`report_dir`	A character scalar of reporting.
`norm_methods`	A character vector of methods, including raw, illumina, swan, quantile, noob, funnorm, yamat, dkfz, quantile, methylcnv. Default to all of them.
`batch`	factor or vector indicating batches. Default to `NULL`, - do not remove batch effect.
`batch2`	optional factor or vector indicating a second series of batches. Default to `NULL`, - do not remove batch effect.
`pipe_file`	A character scalar of the filename storing the pipeline result sample by sample. Turn off the saving by set it to `NULL`. Default to `NULL`.
`dnacopy_seed`	An integer scalar to set the seed. Default to 1.
`segment_grid`	A `data.frame` of parameters for `CNV.segment`. If it is not set, generate one use `.conumee_segment_grid`.
`overwrite`	A logical scalar. Default to FALSE.
`verbose`	A logical scalar. Default to TRUE.
`...`	Any arguments passed to `CNV.segment`.

The ref and qry should include at least the following columns Sample_Name, Gender, Batch, Sample_Prep. Batch effects are removed on methylation and unmethylation signals (log2 transformed) separately. The function outputs the result in a directory. Each subdirectory is a normalzation method. A deeper subdirectory is named in a pattern of "batch-batch2". Then, it contains the subdirectories of sample names, in which there is a segment table named "segments.csv".

An object of ConumeePipe class.

markgene/yamatCN documentation built on Dec. 7, 2019, 4:36 a.m.