GCMS correction Files

“midcor.R” is an “R”-program performs a primary analysis of isotopic isomers (isotopomers) distribution obtained by Gas Cromatography coupled with Mass Spectrometry (GCMS). The aim of this analysis is to have a correct distribution of artificially introduced isotopes. To this end the program performs a correction for natural occurring isotopes and also correction for “impurities” of the assay media that give peaks overlapping with the those produced by the artificially introduced label.
This program offers two ways of corrections of “impurities” resulted from overlapping the assayed mass isotopomer distribution with peaks produced either by unknown metabolites in the media, or by different fragments produced by the assayed metabolite.

How to work with the program.
1. Open terminal, change directory to the working directory containing the file with the R-program (“midcor.R”) and files with input data.
2. Enter into the R shell:
???@???:~/R$ R
3. Read the R-script:
> source("midcor.R")
4. Read a particular input file with data designed for analysis and analyze them using the functions available in Midcor:
> correct("filename")
Here "filename" is the name of the input file contaiming raw GCMS data
The input data is a file with raw mas/charge (m/z) distribution provided directly by a GCMS machine. An example of input data file is shown below. Comments are included between *<>*, they should not be present in a real input file.

*<content of an input file is below>*
carbons_total 18            *<Total number of carbons in the derivated fragment>*
fragment 4                     *<number of carbons in the assayed metabolite (or its fragment) >*
silicio 3                         *<Total number of Si atoms in the derivated fragment>*
m/z                  	417	418	419	420	421	422	423 
R29_01.D             	911	144256	52288	24832	9517	2647	1037 
R29_02.D             	929	127872	46416	21664	7421	2285	688 
R29_03.D             	1039	135296	49192	22760	7967	2202	681 
B29_01.D             	782	109024	39368	18592	7084	1894	674 
B29_02.D             	1059	134528	49200	22968	7643	2027	644 
B29_03.D             	1177	151808	54456	25152	8434	2396	773 
B10n8_01.D             	867	105080	40672	29472	22152	68248	21936 
B10n8_02.D             	1294	148416	56784	41776	29216	95936	31608 
B10n8_03.D             	1311	150528	57464	42104	28776	95184	30776 
….............................................................................................................................
*<whatever number of lines, where the first word is the name of the sample. First three characters of this name, if they are the same, correspond to various injections of the same sample (). Next two characterizes should be the same for various samples corresponding to the same conditions of incubation. The subsequent values correspond to the given m/z in the GCMS spectrum.
Below is the last line, recognized by “fin” in the beginning.   >*
finR24_02.D            	1381	101072	41633	33811	22779	53581	17014
*<end of an input file>*


The function “correct” reads the user provided GCMS data file, normalizes the input GCMS data, so that the sum of m/z values equals to 1, corrects the obtained distribution for the presence of naturally occurring isotopes, and, for the case of commercial unlabeled sample, calculates the difference between the corrected distribution and expected one. Then it uses the obtained difference to ultimately correct for impurity the samples containing metabolites of living cells in the presence of labeled substrates.
The results of such corrections are saved into the output file, which name is the name of input file with “_c” added at the end.

Below an example of output data file is shown. First it prints normalized data corrected for natural isotope occurrence calculated for each row of the input file (last column shows the sum of all m/z fractions given in a corresponding row):

1 R29_01.D 0.9601  0.009513 0.009890 0.020228  2.902e-04 0 1 
2 R29_02.D 0.9664  0.010092 0.007329 0.014540  1.663e-03 0 1 
3 R29_03.D 0.9672  0.010683 0.006033 0.015567  5.453e-04 0 1 
4 B29_01.D 0.9630  0.008246 0.008875 0.019958 -9.602e-05 0 1 
5 B29_02.D 0.9663  0.012697 0.007689 0.013101  1.716e-04 0 1 
6 B29_03.D 0.9740  0.006081 0.005272 0.013777  8.850e-04 0 1 
7 B10n8_01.D 0.6161  0.020621 0.064714 0.067461  2.311e-01 0 1 

Then it prints the data corrected for impurities after grouping together the counts for various injections referred to the same sample,  and mean and standard deviation, calculated for the injections referred to the same samples:
R29_01.D 	1.000000  4.903e-14  8.902e-15  4.065e-14 -3.297e-14
B29_01.D 	1.004161 -1.229e-03 -7.621e-04 -1.751e-03 -4.178e-04
**mean** 	1.000000  0.000e+00  0.000e+00  0.000e+00  0.000e+00
**sd** 		0.002942  8.692e-04  5.389e-04  1.238e-03  2.954e-04

An example of input data can be found in the file “E188Aspartate.txt”, and the corresponding output data are in the file “E188Aspartate.txt_c”. The theoretical basis for corrections used in "midror.R" are described in the file "supplement.doc".