combine_peaklists: Read and Combine Bruker MSI peaklist files.
In Armadilloa16/dipps: Tools for Bruker Peaklist Format MSI Data

Description Usage Arguments Details Value Functions Assumptions See Also Examples

View source: R/io.R

Peakpicking using Bruker mass spectrometry imaging software generally produces a folder filled with many peaklists — one for each spectrum. combine_peaklists reads all such peaklist files in a given folder, summarises and writes the relevant information to two tables. The load_* functions are for reading the tables created by combine_peaklists.

load_peaklist(o.name, o.path = ".")

load_speclist(o.name, o.path = ".")

combine_peaklists(i.path, i.name = "peaklists", o.path = ".",
  o.name = NULL)

`o.name`	Name identifying the dataset of interest. If left as `NULL` for `combine_peaklists`, this will default to `basename(i.path)`, which is then what should be passed to the `load_*` functions.
`o.path`	Path to where output files should be written.
`i.path`	Path to dataset of interest. This will usually be a folder with a subfolder (`i.name`) that contains the peaklist files.
`i.name`	Name of the subfolder in `i.path` that contains the peaklist files.

Typically, Bruker raw data will be stored all in a single folder (often with the ".d" extension), with a folder name that identifies the relevant information about the run. The only compulsory argument to combine_peaklists is a filepath to that folder, i.path under arguments. Typically the peaklist files themselves will be in a subfolder of this folder, often called "peaklists" (can be specified via i.name). This is the only information with which combine_peaklists is concerned — any other files in said folder are ignored. The output files will by default be named using the folder name provided in i.path, but this can be overridden by supplying o.name. These output files are by default created in the current working directory but this can be changed via o.path.

combine_peaklists creates and writes relevant information to two tables: a peak-list, and a spectrum-list:

Peak-list: A table in which each row corresponds to a peak. Consists of all original peaklists concatenated together with one additional column, Acq, identifying the spectrum from which a peak originated.
Spectrum-list: A table in which each row corresponds to a spectrum. Columns contain information relevant at a spectrum level, specifically this table contains five columns total:
- fname: The original peaklist filename.
- R: The region numbers of the corresponding spectra.
- X: The X-coordinates of the corresponding spectra.
- Y: The Y-coordinates of the corresponding spectra.
- Acq: An integer identifier used to cross-reference to the peak-list table.

Note that Acq will correspond to the order of acquisition of the spectra if the spectra where acquired in increasing order of first region number, second Y-coordinate and third X-coordinate, in that order of priority. This order or acquisition is a common default on Bruker Flex instruments.

On successful completion returns the number of empty spectra found — peaklist files with a header but no peaks. If no peaklist files are found at all returns -1 and a warning.

load_peaklist: Reads Peak-list Produced by combine_peaklists.
load_speclist: Reads Spectrum-list Produced by combine_peaklists.

combine_peaklists assumes that the data is in a particular format – that produced by Bruker software peakpicking at the time of writing this vignette, specifically that:

Each spectrum is represented by a single peaklist text-file containing any peaks detected in that spectrum.
All peaklist files (for any given run) are stored in the same folder.
The filename of each peaklist file contains a match to the perl regular expression R(?P<r>\d{2,3})X(?P<x>\d{3,4})Y(?P<y>\d{3,4}) in which the named capture groups 'r', 'x', and 'y', correspond to the region number, x-coordinate, and y-coordinate of the corresponding spectrum.
The contents of each peaklist file is formatted such that it can be correctly read with the use utils::read.table(. , header = TRUE). TODO: Update this to be more specific and instead use base::scan, and remove the load_* functions instead replacing them with suggestions for direct use of a function such as utils::read.csv.

Note that combine_peaklists checks for duplicate spectra (peaklist files with the same region number, X-coordinate, and Y-coordinate) and will throw an error if it finds any.

basename