Version: 0.1.1
Compares specific metabolite levels in two NMR spectra of blood serum/plasma samples.
This tool implements a comparison between the NMR spectra of a reference sample and a test sample, based on selected metabolite signals.
Two aliquots of the same sample have the same metabolomic profiles, and thus have exactly the same NMR spectra. Based on this premise, the NMR spectrum can be used to check the quality of samples.
This tool compares the NMR spectrum of a blood serum/plasma sample with the NMR spectrum of a different aliquot of the same sample acquired at a different time (e.g., spectrum acquired at the moment of the sample collection vs. spectrum acquired after long-term storage). The metabolites tested are those ones that were experimentally identified as "sensible" indicators of sample quality. The respective NMR signals are selected, and their relative concentrations inside each of the samples compared are measured. Spectral total area is also calculated to have information about the presence of contaminations. The operation is part of a QC procedure for biobanks, where samples are periodically analyzed to detected and measure long-term quality decay.
The tool generates a table and a set of plots indicating the signal differences and whether they are significant.
- NMR Processing
- NMR
- Metabolomics / Targeted
- Leonardo Tenori (CERM)
- Veronica Ghini (CERM)
- Antonio Rosato (CERM)
- Luca Pireddu (CRS4)
- Luca Pireddu (CRS4)
For local individual installation:
docker pull docker-registry.phenomenal-h2020.eu/phnmnl/nmr-integrals
For direct Docker usage:
docker run docker-registry.phenomenal-h2020.eu/phnmnl/nmr-integrals
Usage: ./integrals.R [options] REFERENCE_SPECTRUM TEST_SPECTRUM
Options:
--left=N
Left
--right=N
Right
--where=N
Where
--plotfile=FILE.PDF
Plot file
--metabolites
Metabolites table in tab-separated format
-h, --help
Show this help message and exit
REFERENCE_SPECTRUM
Reference dataset (Bruker monodimensional 1H cpmg NMR spectrum)
TEST_SPECTRUM
Test dataset (Bruker monodimensional 1H cpmg NMR spectrum)
Left - Left margin of the integration range (ppm value)
Right - Right margin of the integration range (ppm value) (must be greater than left)
Where - ppm value (inside the integration rage) where we want to align the signals of the two spectra.
Reference spectrum - Bruker monodimensional 1H cpmg NMR spectrum
Test spectrum - Bruker monodimensional 1H cpmg NMR spectrum
Signal Metabolites - Optional table of signal metabolites, to override default behaviour.
The metabolites table specifies, for each selected metabolite, an integration window for one or more of its signals inside the spectra. The program will by default use a table (shown below) made specifically for the purpose of detecting decay of blood serum/plasma samples. You may however customize this table to apply the technique to other types of samples.
Here is the table format.
Columns:
1st column - Metabolite name
2nd column - Right margin of the integration range (ppm value)
3rd column - Left margin of the integration range (ppm value)
4th column - ppm value (inside the integration rage) where we want to align the signals of the two spectra.
5th column - base line correction (0: no /1: yes)
6th column - signal alignment at where (0: no /1: yes)
The table below is the default table, which has been verified to return sensible values:
Metabolite right ppm left ppm ppm align baseline correct on/off alignment on/off
totalarea 1.3000 0.9000 1.2797 0 1
totalarea1 4.2 1.3000 1.3246 0 1
totalarea2 9.0 5.00 5.2431 0 1
glucose 5.2500 5.2200 5.2367 1 1
lactate 4.1137 4.0920 4.1085 1 1
citrate 2.5600 2.5400 2.55 1 1
pyruvate 2.3780 2.3650 2.373 1 1
ethanol 1.2028 1.1926 1.1969 0 0
ethanol1 1.1913 1.1798 1.1885 0 0
glycine 3.5700 3.5583 3.563 1 1
acetate 1.9231 1.9165 1.9189 1 1
lypids 1.3150 1.2500 1.278 1 1
glycerol 3.6959 3.6368 3.6674 0 0
- Plots of ppm vs relative intensity, for each metabolite (PDF format).
- Table (tab-separated columns) with the difference in the relative concentration of the selected metabolites. Threshold: +/-20%.
Table columns:
1st column - Metabolite name
2nd column - ok/bad
4th column - relative concetration difference (percentage)
The 3rd column is always "diff" and the 5th one is always a "%" sign.
This work was done in collaboration with SPIDIA4P.