groupPEAKS: Get peak-groups representing different chemical spectra in...
In lauzikaite/massFlowR: LC-MS Data Pre-Processing

Description Usage Arguments Details Value

Function performs peak-picking and peak-grouping for each LC-MS datafile independently.

1	groupPEAKS(file = NULL, out_dir = NULL, cwt = NULL, ncores = 2, thr = 0.95)

`file`	`character` with path to metadata csv file, which must incluce columns 'filename', 'run_order' and 'raw_filepath'.
`out_dir`	`character` specifying desired directory for output.
`cwt`	`CentWaveParam` class object with parameters for centWave-based peak-picking.
`ncores`	`numeric` defining number of cores to use for parallelisation. Default set to 1 for serial implementation.
`thr`	`numeric` defining correlation coefficient threshold, above which peak pairs will be considered as correlated. Default set to 0.95.

Function performs peak-picking using the centWave algorithm from package xcms. Picked-peaks are then grouped into chemical spectra. For each peak in the sample:

Co-eluting peaks are found.
EIC correlation between all co-eluting peaks is performed.
Network of peaks with high EIC correlation (with coefficients above the selected threshold) is built.

Co-eluting peaks within a network of high EIC correlation originate from the same chemical compound and therefore form a chemical spectrum.

For each LC-MS file, function writes a table with picked-peaks and their peak-groups into separate files, in the defined out_dir directory. Inputed metadata file is updated to include a column 'proc_filepath', specifying full path to processed csv files peak-group tables.

lauzikaite/massFlowR documentation built on April 29, 2020, 9:45 a.m.