3.21. Metabolome 2022

Dataset category

Metabolome

Summary

Metabolome analysis results derived from approximately 53,000 Japanese individuals

References

  • Koshiba et al. [9]

  • Saigusa et al. [10]

  • Saigusa et al. [11]

Samples analyzed

Plasma

Number of samples and analysis platform

Category

Analysis platform

Nunber of samples

NMR
NMR (Bruker: 600MHz),
CryoProbe SampleJet
54,956
(Non pregnant: 40,529, Pregnant: 14,427, Repeat assessment: 4,599)

LC-MS/G-Met v1
C18 column: UHPLC-Q-TOF/MS (Waters: Synapt G2-Si),
HILIC column: HPLC-Q-FT/MS (Thermo Fisher Scientific: QExactive)

1,266

LC-MS/G-met v2

HPLC-Q-FT/MS (Thermo Fisher Scientific: QExactive)

2,972

LC-MS/T-Met

UHPLC-MS/MS (Thermo Fisher Scientific: TSQ Quantiva)

2,365

LC-MS/kit180

UHPLC-MS/MS (Waters: Xevo TQ-S)

1,483 (Repeat assessment: 579)

LC-MS/kit500

UHPLC-MS/MS (Waters: Xevo TQ-S, TQ-XS)

9,231 (Repeat assessment: 1,338)

GC-MS/T-Met

GC-MS/MS (Shimadzu: TQ8040)

4,611 (Repeat assessment: 646)
Age / BMI distributions (non pregnant)

Category

Age distribution

BMI distribution

All

 ../../_images/age-sex.dodge.jMorp2022.all.nonPregnant.n40529.png ../../_images/BMI-sex.dodge.jMorp2022.all.nonPregnant.n40529.png

NMR

 ../../_images/age-sex.dodge.NMR__TCN.nonPregnant.n40529.png ../../_images/BMI-sex.dodge.NMR__TCN.nonPregnant.n40529.png

LC-MS/G-Met v1

 ../../_images/age-sex.dodge.LC-MS.G-Met.v1__TCOZ.nonPregnant.n1266.png ../../_images/BMI-sex.dodge.LC-MS.G-Met.v1__TCOZ.nonPregnant.n1266.png

LC-MS/G-met v2

 ../../_images/age-sex.dodge.LC-MS.G-Met.v2__TCI.nonPregnant.n2972.png ../../_images/BMI-sex.dodge.LC-MS.G-Met.v2__TCI.nonPregnant.n2972.png

LC-MS/T-Met

 ../../_images/age-sex.dodge.LC-MS.T-Met__TCL.nonPregnant.n2365.png ../../_images/BMI-sex.dodge.LC-MS.T-Met__TCL.nonPregnant.n2365.png

LC-MS/kit180

 ../../_images/age-sex.dodge.LC-MS.T-Met.Kit180__TCB.nonPregnant.n1483.png ../../_images/BMI-sex.dodge.LC-MS.T-Met.Kit180__TCB.nonPregnant.n1483.png

LC-MS/kit500

 ../../_images/age-sex.dodge.LC-MS.T-Met.Kit500__TCM.nonPregnant.n9231.png ../../_images/BMI-sex.dodge.LC-MS.T-Met.Kit500__TCM.nonPregnant.n9231.png

GC-MS/T-Met

 ../../_images/age-sex.dodge.GC-MS.T-Met__TCS.nonPregnant.n4611.png ../../_images/BMI-sex.dodge.GC-MS.T-Met__TCS.nonPregnant.n4611.png
Rule of metabolite ID convension

Metabolites in this dataset are assigned metabolite IDs in the form of TCx123456. The first two characters (TC) stands for ToMMo Compound ID (TC-ID). The third letter indicates how the metabolite is measured, and is divided as follows:

Prefix

Meaning

TCN

NMR

TCZ

LC-MS G-Met metabolome in HILIC mode ver.1 using HPLC

TCI

LC-MS G-Met metabolome in HILIC mode ver.2 using UHPLC

TCO

LC-MS G-Met metabolome in C18 mode using UHPLC

TCL

LC-MS T-Met metabolome

TCB

LC-MS T-Met metabolome in kit180

TCM

LC-MS T-Met metabolome in kit500

TCS

GC-MS T-Met metabolome

The last six digits are numbers assigned to each metabolite within each data source. In MS metabolome, data sources can be further divided into positive and negative modes. A number less than 500,000 indicates that the metabolite was detected in negative mode; otherwise, it is a metabolite detected in positive mode.

Automatic quantification of metabolites

The concentrations of metabolites were automatically estimated from NMR spectra by using several regression models. Based on more than 1,000 concentration data carefully calculated by experts, both linear regression model and neural network model were built for each metabolite. We selected the model with best performance by using R-squared (R2) values as an evaluation index. We provide a reliability score of estimated concentration on a four-tiered scale: “Triple Stars (★★★)”, “Double Stars (★★☆)”, “Single Star (★☆☆)” and “Zero Star (☆☆☆)”. Each category corresponds to R2 value >=0.9, >=0.7, >=0.6, and <0.6, respectively. Outliers were defined by this protocol as those concentration is >10 SD after automatically estimation and were excluded in each compound.

Note related to Hypoxanthine and Inosine

Because values of metabolites, (TCN000044, TCI006689, TCS000091, TCO501589, TCZ000947) and Inosine (TCN000045, TCI010703), strongly depend on the stored time until specimen processing for banking, values of the samples that were not processed on the day were excluded for these two metabolites.

Related pages on jMorp website