fiems_ltq_main: LTQ Mass Binning

In tonedivad/FIEmspro: Flow Injection Electrospray Mass Spectrometry Processing: \\ data processing, classification modelling and variable selection in metabolite fingerprinting

Description

Main Routine for ‘Mass Binning’ to nominal mass and ‘Mass Spectrum’ generation in high-throughput Flow Injection Electrospray Ionisation Mass Spectrometry (FIE-MS). This routine reads ANDI NetCDF files (*.cdf) of LTQ *.raw data files converted in the Xconvert program (Xcalibur, Thermo Finnigan).

Usage

fiems_ltq_main(my_path,runinfo, y1,y2,y3,y4,limit=0.7,
               itp=65537,itn=131073,hrng=50,lrng=15,
               save.file=TRUE,file.name="LTQ-mean.RData")

Arguments

my_path	A character string indicating the working directory where runinfo.csv file and folder containing *.cdf-files are located.
runinfo	A *.csv file containing at least the following run information (header row): pathcdf and filecdf . For details, see the description in Examples below.
y1	A numeric value used for mass spectrum generation: start scan ‘sample’. For details, see the description in Examples below.
y2	A numeric value used for mass spectrum generation: end scan ‘sample’. For details, see the description in Examples below.
y3	A numeric value used for mass spectrum generation: start scan ‘background’. For details, see the description in Examples below.
y4	A numeric value used for mass spectrum generation: end scan ‘background’. For details, see the description in Examples below.
limit	A numeric value defining the rounding limit for binning m/z-values to nominal mass.
itp	itp is an identifier for Scan Type ‘full’ and Data Type ‘centroid’ (Instrument Setup, Xcalibur) for data scans acquired in positive ionisation mode using the LTQ (Default=65537).
itn	itp is an identifier for Scan Type ‘full’ and Data Type ‘centroid’ (Instrument Setup, Xcalibur) for data scans acquired in negative ionisation mode using the LTQ (Default=131073).
hrng	hrng is an identifier for the high mass range. The argument has to match the ‘First Mass (m/z)’ of the scan range used for acquiring data using the LTQ ‘Instrument Setup -> Mass Range: normal’.
lrng	lrng is an identifier for the low mass range. The argument has to match the ‘First Mass (m/z)’ of the scan range used for acquiring data in LTQ ‘Instrument Setup -> Mass Range: low’.
save.file	A logical value indicating whether or not to save the results (default is TRUE).
file.name	A character for saved file name if save.file is TRUE.

Details

This routine is designed to handle four Scan Events (Xcalibur, Thermo Finnigan). Each *.cdf-file will result four mass spectra. Principle in brief: load *.cdf-file (pathcdf and filecdf information in runinfo.csv); sort scans (m/z and intensity values) into four lists. For each list: bin m/z-values to nominal mass between ‘limit - 1’ and ‘limit’; sum up intensities of binned m/z values; generate sample matrix ‘smat’ between scans y1 and y2 and background matrix ‘bmat’ between scans y3 and y4; subtract: mat=smat-bmat; calculate mean of resulting matrix ‘mat’; potential negative values are set to ‘zero’. The implemented timer-function should be accurate for up to 24 hours which could comprise more than 7000 *.cdf-files per experiment. See the description in the Examples below.

Value

A list containing the following components:

posh	Matrix [runs x nominal masses] of high mass range in positive ionisation mode.
posl	Matrix [runs x nominal masses] of low mass range in positive ionisation mode.
negh	Matrix [runs x nominal masses] of high mass range in negative ionisation mode.
negl	Matrix [runs x nominal masses] of low mass range in negative ionisation mode.
runinfo	Same as argument stored for reference purposes. Additional information for each run like sample name or class can be used for further analysis (e.g. nlda).
scrng	A vector of y1, y2, y3 and y4 stored for reference purposes.
limit	Same as argument stored for reference purposes.

Note

The returned values are saved by default as LTQ-mean.RData in folder my_path. Additionally, single items are saved by default as TEXT files: posh.txt, posl.txt, negh.txt, negl.txt, myparam.txt (containing scrng and limit for reference purposes).

Author(s)

Manfred Beckmann [e-mail]

See Also

fiems_lct_main

Examples

## To run fiems_lqt_main, copy and paste the following code segment. Uncomment
## and change the file path and name appropriately.


## Example profiles can be downloading on the FIEmspro webpage
## 050509-Abr1.zip must be extrated in folder that defines 'my_path'

## For e.g.
## Not run: my_path <- "D:/Temp/050509-Abr1"
## The same folder should also contain a 'runinfo' file
## e.g.
## Not run: runinfo <- "runinfo.csv"
## Process each profile defined in 'runinfo'
## Not run: tmp <- fiems_lct_main(my_path,runinfo,35,95,190,250,limit=0.82,
               save.file=TRUE,file.name="LTQ-mean.RData")
## End(Not run)

## ===================================================================
## Arguments and matrices are saved in 'my_path', ideally the working 
## directory of the experiment. For explanations regarding input 
## arguments see below ...

## required is a file named by default 'runinfo.csv'
## (comma separated variables, generated in e.g. MS-Excel)
## the structure should be the following to ease data pre-processing:

##    A     |             B                   |    C       |   D
## injorder | pathcdf                         | filecdf    | batch
##-----------------------------------------------------------------
##  1       | D:/../070122-ABR1-A-repeat/cdf  | 01.cdf     | 1
##  2       | D:/../070122-ABR1-A-repeat/cdf  | 02.cdf     | 1
## and so on...

##  Columns:
## 'injorder' is injection order of samples (good for investigating drifts)
## 'pathcdf'  is path of folder containing "*.cdf"-files. Each run-sequence
##            or batch of runs might have its own folder.
## 'filecdf'  is the actual filename of an "*.cdf"-file.
## 'batch'    is the number of the batch the run belongs to (good for
##            investigating batch effects)

## In practice the file will contain further information regarding sample name,
## class/group information and probably other meta-data describing a 
## sample.

## LTQ Instrument Method for Flow-Injection-ESI-MS (FIE-MS):
## - 1 Segment, 5 min Acquisition
## - 4 Scan Events:
##   -- 1: ITMS + c norm o(50.0-2000.0)
##   -- 2: ITMS + c low injrf=20.0 o(15.0-200.0)
##   -- 3: ITMS - c norm o(50.0-2000.0)
##   -- 4: ITMS - c low injrf=20.0 o(15.0-200.0)

## Infusion Profile (Sketch):
##           _
##          / \
##         /   \
##        /     \_
##       /        \______
## _____/                \________________________
## 0       1        2        3         4         5 [min]
##       |--- ---|               |-------|
##      [x1]   [x2]            [x3]    [x4]   [scan reading]
##        sample                background
## Using the above LTQ Instrument Method for FIE-MS
##   the actual scan readings x1 to x4 of e.g. scan event 1 have to be
##   subtracted by 1 (the Scan Event) and
##   divided by 4 (total of 4 Scan Events):
##      e.g.  [y1] = ([x1]-1)/4  =>  scrange = c(y1,y2,y3,y4)
##      with (ideally):  y2 - y1 = y4 - y3

## Raw data conversion to ANDI NetCDF-file: 
##      XConvert-program (Xcalibur, Thermo-Finnigan)

tonedivad/FIEmspro documentation built on May 31, 2019, 6:20 p.m.