Research ArticleMETABOLOMICS

Urinary metabolic signatures of human adiposity

See allHide authors and affiliations

Science Translational Medicine  29 Apr 2015:
Vol. 7, Issue 285, pp. 285ra62
DOI: 10.1126/scitranslmed.aaa5680
  • Fig. 1. Schema of study design.

    Individuals were recruited into the study from general and occupational population samples in the United States and UK. Each individual made four clinic visits, the first two on consecutive days and the second two on consecutive days 3 weeks later. At each visit, blood pressure was measured twice, and a complete 24-hour dietary recall was obtained by trained interviewers. Data on height, weight, and extensive questionnaire information were also obtained. A timed 24-hour urine collection was commenced at the first (and third) visit and completed at the second (and fourth) visit the following day. Each 24-hour urine collection was mixed together, and urine aliquots were obtained from the pooled urine sample. Urinary aliquots from individuals with complete data were measured with 1H NMR spectroscopy.

  • Fig. 2. Associations of BMI with urinary metabolites in the U.S. INTERMAP cohort (n = 1880).

    Urine was collected for 24 hours from individuals on two separate occasions 3 weeks apart. Each individual completed four 24-hour dietary recalls, and eight blood pressure measurements and questionnaire information were obtained. The partial correlation between each 1H NMR variable and BMI was adjusted for age, gender, and population sample (model 1). (A) Average 600 MHz 1H NMR spectrum of the first urine collection. (B) Manhattan plot showing −log10(Q) × sign of partial correlation for each of the 7100 spectral variables. Significance of 1H NMR was determined based on a Q value threshold of 1%; in addition to this, both adjacent variables must also pass Q ≤ 0.01 and must have the same sign. Statistically significant peaks are colored red if directly associated with BMI and blue if inversely associated. Metabolites significantly associated with BMI (numbered according to spectral position) were as follows: 1: ketoleucine, 2: leucine, 3: valine, 4: 2-hydroxyisobutyrate, 5: alanine, 6: lysine, 7: N-acetyl signals from urinary glycoproteins, 8: N-acetyl neuraminate, 9: phenylacetylglutamine, 10: glutamine, 11: proline betaine, 12: 4-cresyl sulfate, 13: succinate, 14: citrate, 15: dimethylamine, 16: TMA, 17: dimethylglycine, 18: creatinine, 19: ethanolamine, 20: O-acetyl carnitine, 21: glucose, 22: 3-methylhistidine, 23: glycine, 24: hippurate, 25: pseudouridine, 26: NMNA, 27: 3-hydroxymandelate, 28: tyrosine, 29: 4-hydroxymandelate, 30: formate, U1 to U26 unidentified metabolites. These data are tabulated in Table 1 and table S4. Some signals overlap; however, they were unequivocally identified using various statistical and experimental methods (table S2).

  • Fig. 3. Multicompartmental metabolic reaction network illustrating human adiposity–related urinary metabolic signatures in the U.S. population (n = 1880).

    Metabolites are included that passed the ST-FDR threshold of Q ≤ 0.01 for both models 1 and 2 (1H NMR metabolites, see Table 1) and the Bonferroni threshold of P ≤ 4.55 × 10–4 for both models 1 and 2 (metabolites detected by IEC, see Table 2). The background shading illustrates different types of metabolism based on closest affinity classification. Table S10 lists the full names and abbreviations for the metabolites. Table S11 lists the closest affinity classifications for metabolites shown in this figure. Dotted lines indicate the closest related metabolite in the network for metabolites that are not listed in the KEGG database, based on available literature.

  • Table 1. Structurally identified 1H NMR–derived urinary metabolites associated significantly with BMI in 1880 U.S.

    INTERMAP participants using first urine collection specimens. Excluding metabolic outliers based on Hotelling’s T2 test (n = 132) and participants with doctor-diagnosed diabetes mellitus (n = 152). Partial correlation (r) and corresponding P values are listed for each metabolite. Statistical significance is based on the ST-FDR, Q ≤ 0.01 for each spectral variable and the two adjacent spectral variables.

    Urinary variableModel 1*Model 2*Model 3*†‡
    rPQrPQrPQ
    Urinary glycoproteins0.292.04 × 10–364.15 × 10–330.279.63 × 10–333.22 × 10–290.273.09 × 10–331.23 × 10–29
    Ketoleucine–0.227.19 × 10–223.56 × 10–19–0.201.45 × 10–188.07 × 10–16–0.236.84 × 10–236.81 × 10–20
    Tyrosine + 4-hydroxymandelate§0.211.35 × 10–195.72 × 10–170.203.64 × 10–181.74 × 10–150.173.60 × 10–148.97 × 10–12
    Succinate–0.203.22 × 10–189.56 × 10–16–0.151.19 × 10–108.44 × 10–9–0.142.42 × 10–91.80 × 10–7
    Glutamine–0.196.05 × 10–171.50 × 10–14–0.151.38 × 10–109.07 × 10–9–0.131.70 × 10–88.78 × 10–7
    N-acetyl neuraminate0.192.02 × 10–164.07 × 10–140.195.25 × 10–172.20 × 10–140.223.75 × 10–222.99 × 10–19
    3-Methylhistidine0.182.87 × 10–155.01 × 10–130.172.79 × 10–133.90 × 10–110.125.68 × 10–71.18 × 10–5
    Citrate–0.188.61 × 10–151.16 × 10–12–0.151.54 × 10–109.86 × 10–9–0.143.45 × 10–92.41 × 10–7
    Hippurate–0.181.52 × 10–141.77 × 10–12–0.143.39 × 10–101.86 × 10–8–0.121.24 × 10–74.02 × 10–6
    Creatinine0.178.88 × 10–146.28 × 10–120.171.78 × 10–132.84 × 10–11
    NMNA–0.169.04 × 10–134.55 × 10–11–0.141.15 × 10–94.31 × 10–8–0.128.60 × 10–83.14 × 10–6
    Leucine0.161.16 × 10–125.45 × 10–110.161.03 × 10–111.01 × 10–90.106.63 × 10–67.03 × 10–5
    Pseudouridine0.162.10 × 10–128.53 × 10–110.144.09 × 10–102.11 × 10–80.143.21 × 10–92.29 × 10–7
    Lysine0.163.67 × 10–121.38 × 10–100.145.23 × 10–102.48 × 10–80.134.39 × 10–81.90 × 10–6
    Formate–0.164.50 × 10–121.65 × 10–10–0.133.17 × 10–85.74 × 10–7–0.094.84 × 10–53.28 × 10–4
    2-Hydroxyisobutyrate0.167.24 × 10–122.29 × 10–100.114.89 × 10–63.25 × 10–50.097.60 × 10–54.59 × 10–4
    4-Cresyl sulfate–0.158.09 × 10–111.57 × 10–9–0.151.32 × 10–109.01 × 10–9–0.153.22 × 10–103.13 × 10–8
    Glycine–0.146.29 × 10–108.82 × 10–9–0.122.91 × 10–73.22 × 10–6–0.112.12 × 10–63.09 × 10–5
    TMA0.142.04 × 10–92.49 × 10–80.136.41 × 10–91.66 × 10–70.111.52 × 10–62.46 × 10–5
    Ethanolamine0.142.72 × 10–93.22 × 10–80.133.95 × 10–86.65 × 10–70.133.17 × 10–81.45 × 10–6
    Glucose0.143.27 × 10–93.84 × 10–80.133.21 × 10–85.76 × 10–70.126.85 × 10–82.66 × 10–6
    Proline betaine–0.137.29 × 10–97.96 × 10–8–0.101.04 × 10–56.10 × 10–5–0.091.88 × 10–49.46 × 10–4
    Phenylacetylglutamine–0.131.39 × 10–81.37 × 10–7–0.121.79 × 10–72.21 × 10–6–0.111.15 × 10–61.97 × 10–5
    Dimethylamine0.131.74 × 10–81.69 × 10–70.125.05 × 10–75.09 × 10–60.096.43 × 10–54.07 × 10–4
    3-Hydroxymandelate–0.132.81 × 10–82.57 × 10–7–0.122.72 × 10–73.05 × 10–6–0.101.98 × 10–51.68 × 10–4
    Dimethylglycine0.121.43 × 10–71.04 × 10–60.111.18 × 10–61.01 × 10–50.072.98 × 10–38.49 × 10–3
    Valine0.122.39 × 10–71.62 × 10–60.108.15 × 10–64.97 × 10–50.081.20 × 10–34.02 × 10–3
    Alanine0.123.32 × 10–72.18 × 10–60.112.29 × 10–61.75 × 10–50.094.72 × 10–53.21 × 10–4
    O-acetyl carnitine0.101.46 × 10–55.37 × 10–50.101.38 × 10–57.62 × 10–50.032.41 × 10–12.48 × 10–1
    Ketoleucine/leucine ratio–0.248.85 × 10–265.84 × 10–23–0.231.02 × 10–237.61 × 10–21–0.211.28 × 10–206.80 × 10–18

    *Model 1 is adjusted for age, gender, and sample.

    †Model 2 is adjusted for all factors in model 1 plus cardiovascular disease history (heart disease/stroke), physical activity, medication for hypertension, prescribed lipid-lowering drugs, NSAID use, dietary supplement use, special diet, smoking, education, and total energy intake per day (kcal/day).

    ‡Model 3 is adjusted for all factors in models 1 and 2 plus 24-hour urinary creatinine.

    §The signal at δ 6.91 of tyrosine also includes a signal from its metabolite 4-hydroxymandelate.

    • Table 2. Association with BMI of a set of metabolites measured by targeted IEC (log10 values) in 1880 U.S. INTERMAP participants.

      Excluding metabolic outliers based on Hotelling's T2 test (n = 132) and participants with doctor-diagnosed diabetes mellitus (n = 152). Partial correlation (r) and corresponding P values are listed for each metabolite. Statistical significance based on a Bonferroni threshold of P ≤ 4.55 × 10−4 (P ≤ 0.01/22). n.s., not significant.

      Urinary variableModel 1*Model 2*Model 3*†‡
      rPSignificancerPSignificancerPSignificance
      Taurine0.023.68 × 10–1n.s.0.016.09 × 10–1n.s.–0.111.13 × 10–6
      Threonine0.154.44 × 10–110.152.34 × 10–10–0.023.82 × 10–1n.s.
      Serine0.053.06 × 10–2n.s.0.061.58 × 10–2n.s.–0.124.41 × 10–7
      Asparagine0.017.69 × 10–1n.s.0.009.97 × 10–1n.s.–0.117.93 × 10–7
      Glutamine0.101.29 × 10–50.123.34 × 10–7–0.054.41 × 10–2n.s.
      Glycine–0.023.98 × 10–1n.s.0.016.00 × 10–1n.s.–0.106.08 × 10–6
      Alanine0.204.85 × 10–180.183.47 × 10–150.023.65 × 10–1n.s.
      Valine0.196.87 × 10–160.176.51 × 10–140.073.87 × 10–3n.s.
      Cystine0.334.85 × 10–490.304.59 × 10–410.186.80 × 10–15
      Methionine0.061.06 × 10–2n.s.0.054.32 × 10–2n.s.–0.031.57 × 10–1n.s.
      Isoleucine0.124.23 × 10–70.102.11 × 10–50.052.73 × 10–2n.s.
      Leucine0.083.66 × 10–40.067.53 × 10–3n.s.0.025.08 × 10–1n.s.
      Tyrosine0.322.02 × 10–450.308.28 × 10–410.181.58 × 10–14
      Phenylalanine0.208.35 × 10–190.191.42 × 10–160.089.99 × 10–4n.s.
      Ethanolamine0.262.00 × 10–290.272.19 × 10–310.147.39 × 10–10
      Lysine0.274.64 × 10–330.267.81 × 10–300.121.06 × 10–7
      1-Methylhistidine0.119.47 × 10–70.097.20 × 10–5–0.054.48 × 10–2n.s.
      Histidine0.143.02 × 10–90.156.23 × 10–110.016.43 × 10–1n.s.
      Tryptophan0.154.49 × 10–110.151.79 × 10–100.094.35 × 10–5
      3-Methylhistidine0.428.28 × 10–800.381.87 × 10–650.122.14 × 10–7
      Carnosine0.187.16 × 10–150.172.28 × 10–130.022.90 × 10–1n.s.
      Arginine0.088.71 × 10–4n.s.0.072.77 × 10–3n.s.0.031.34 × 10–1n.s.

      *Model 1 is adjusted for age, gender, and sample.

      †Model 2 is adjusted for all factors in model 1 plus cardiovascular disease history (heart disease/stroke), physical activity, medication for hypertension, prescribed lipid-lowering drugs, NSAID use, dietary supplement use, special diet, smoking, education, and total energy intake per day (kcal/day).

      ‡Model 3 is adjusted for all factors in models 1 and 2 plus 24-hour urinary creatinine.

      • Table 3. Association with BMI of (A) metabolites identified by 1H NMR and (B) metabolites measured by targeted IEC (log10 values) in 444 UK INTERMAP participants for first urine collection specimens.

        Excluding metabolic outliers based on Hotelling’s T2 test (n = 47) and participants with doctor-diagnosed diabetes mellitus (n = 5). Partial correlation (r) and corresponding P values are listed for each metabolite. Statistical significance is based on the ST-FDR, Q ≤ 0.05 for 1H NMR metabolites identified and analyzed in the U.S. data. Statistical significance for IEC metabolites is based on a Bonferroni threshold of P ≤ 2.50 × 10−3 (P ≤ 0.05/20).

        Urinary variableModel 1*Model 2*Model 3*†‡
        rPQrPQrPQ
        A. 1H NMR spectroscopy
        Urinary glycoproteins0.331.02 × 10–124.66 × 10–90.331.22 × 10–126.17 × 10–90.361.50 × 10–147.25 × 10–11
        Ketoleucine–0.181.49 × 10–42.50 × 10–3–0.142.66 × 10–32.25 × 10–2–0.161.00 × 10–31.08 × 10–2
        Tyrosine + 4-hydroxymandelate§0.251.17 × 10–75.19 × 10–50.258.71 × 10–85.53 × 10–50.239.69 × 10–75.20 × 10–4
        Succinate–0.142.52 × 10–31.76 × 10–2–0.152.19 × 10–31.98 × 10–2–0.135.62 × 10–33.46 × 10–2
        Glutamine–0.142.62 × 10–31.80 × 10–2–0.135.25 × 10–33.52 × 10–2–0.121.49 × 10–26.70 × 10–2
        N-acetyl neuraminate0.181.74 × 10–42.81 × 10–30.204.56 × 10–51.58 × 10–30.211.67 × 10–51.11 × 10–3
        3-Methylhistidine0.137.71 × 10–33.62 × 10–20.135.23 × 10–33.52 × 10–20.088.41 × 10–21.96 × 10–1
        Citrate0.104.45 × 10–21.22 × 10–1–0.112.40 × 10–29.93 × 10–20.112.26 × 10–28.62 × 10–2
        Hippurate–0.242.75 × 10–75.19 × 10–5–0.231.78 × 10–64.43 × 10–4–0.231.80 × 10–66.70 × 10–4
        Creatinine0.173.96 × 10–45.11 × 10–30.181.68 × 10–43.73 × 10–3
        NMNA–0.087.76 × 10–21.75 × 10–1–0.052.84 × 10–14.29 × 10–1–0.043.69 × 10–14.50 × 10–1
        Leucine0.173.73 × 10–44.88 × 10–30.166.13 × 10–48.57 × 10–30.112.23 × 10–28.58 × 10–2
        Pseudouridine0.244.87 × 10–76.38 × 10–50.251.86 × 10–71.05 × 10–40.248.09 × 10–74.89 × 10–4
        Lysine0.324.30 × 10–129.84 × 10–90.297.11 × 10–109.03 × 10–70.245.78 × 10–74.29 × 10–4
        Formate–0.151.72 × 10–31.35 × 10–2–0.144.66 × 10–33.29 × 10–2–0.137.33 × 10–34.11 × 10–2
        2-Hydroxyisobutyrate0.194.82 × 10–51.11 × 10–30.161.20 × 10–31.32 × 10–20.129.74 × 10–35.01 × 10–2
        4-Cresyl sulfate–0.172.22 × 10–43.36 × 10–3–0.182.09 × 10–44.30 × 10–3–0.161.18 × 10–31.19 × 10–2
        Glycine–0.121.04 × 10–24.37 × 10–2–0.144.65 × 10–33.29 × 10–2–0.121.56 × 10–26.87 × 10–2
        TMA0.211.00 × 10–53.59 × 10–40.218.16 × 10–66.50 × 10–40.222.56 × 10–67.29 × 10–4
        Ethanolamine0.135.23 × 10–32.79 × 10–20.135.25 × 10–33.52 × 10–20.167.04 × 10–48.38 × 10–3
        Glucose0.194.97 × 10–51.13 × 10–30.173.71 × 10–46.02 × 10–30.211.47 × 10–51.06 × 10–3
        Proline betaine–0.121.31 × 10–25.14 × 10–2–0.111.73 × 10–27.81 × 10–2–0.111.85 × 10–27.65 × 10–2
        Phenylacetylglutamine–0.151.30 × 10–31.13 × 10–2–0.175.02 × 10–47.46 × 10–3–0.152.14 × 10–31.76 × 10–2
        Dimethylamine0.143.72 × 10–32.27 × 10–20.166.03 × 10–48.48 × 10–30.143.53 × 10–32.51 × 10–2
        3-Hydroxymandelate–0.094.95 × 10–21.31 × 10–1–0.104.21 × 10–21.41 × 10–1–0.071.31 × 10–12.57 × 10–1
        Dimethylglycine0.166.11 × 10–47.00 × 10–30.174.54 × 10–46.94 × 10–30.138.85 × 10–34.66 × 10–2
        Valine0.144.16 × 10–32.47 × 10–20.121.59 × 10–27.35 × 10–20.103.06 × 10–21.03 × 10–1
        Alanine0.211.10 × 10–53.73 × 10–40.225.07 × 10–65.31 × 10–40.195.81 × 10–51.81 × 10–3
        O-acetyl carnitine0.151.31 × 10–31.13 × 10–20.151.67 × 10–31.66 × 10–20.104.68 × 10–21.35 × 10–1
        Ketoleucine/leucine ratio–0.241.90 × 10–75.19 × 10–5–0.247.11 × 10–72.89 × 10–4–0.203.18 × 10–51.41 × 10–3
        rPSignificancerPSignificancerPSignificance
        B. IEC
        Taurine–0.035.88 × 10–1n.s.–0.044.61 × 10–1n.s.–0.121.38 × 10–2n.s.
        Threonine0.061.92 × 10–1n.s.0.061.91 × 10–1n.s.–0.062.28 × 10–1n.s.
        Serine0.019.16 × 10–1n.s.0.017.60 × 10–1n.s.–0.104.06 × 10–2n.s.
        Asparagine0.009.73 × 10–1n.s.0.009.88 × 10–1n.s.–0.097.05 × 10–2n.s.
        Glutamine0.095.75 × 10–2n.s.0.104.08 × 10–2n.s.–0.018.25 × 10–1n.s.
        Glycine–0.027.16 × 10–1n.s.–0.027.28 × 10–1n.s.–0.103.46 × 10–2n.s.
        Alanine0.167.26 × 10–40.167.36 × 10–40.061.83 × 10–1n.s.
        Valine0.237.14 × 10–70.231.29 × 10–60.081.16 × 10–1n.s.
        Cystine0.331.40 × 10–120.314.59 × 10–110.204.52 × 10–5
        Isoleucine0.061.73 × 10–1n.s.0.062.16 × 10–1n.s.–0.035.18 × 10–1n.s.
        Leucine0.263.04 × 10–80.266.30 × 10–80.104.28 × 10–2n.s.
        Tyrosine0.329.21 × 10–120.321.56 × 10–110.224.99 × 10–6
        Phenylalanine0.251.57 × 10–70.258.54 × 10–80.121.40 × 10–2n.s.
        Ethanolamine0.313.44 × 10–110.318.26 × 10–110.168.32 × 10–4
        Lysine0.251.28 × 10–70.245.56 × 10–70.136.45 × 10–3n.s.
        1-Methylhistidine0.128.90 × 10–3n.s.0.087.99 × 10–2n.s.–0.034.86 × 10–1n.s.
        Histidine0.111.85 × 10–2n.s.0.121.39 × 10–2n.s.0.018.26 × 10–1n.s.
        Tryptophan0.219.20 × 10–60.201.80 × 10–50.121.03 × 10–2n.s.
        3-Methylhistidine0.312.98 × 10–110.295.05 × 10–100.089.31 × 10–2n.s.
        Carnosine0.169.36 × 10–40.172.90 × 10–40.018.14 × 10–1n.s.

        *Model 1 is adjusted for age, gender, and sample.

        †Model 2 is adjusted for all factors in model 1 plus cardiovascular disease history (heart disease/stroke), physical activity, medication for hypertension, prescribed lipid-lowering drugs, NSAID use, dietary supplement use, special diet, smoking, education, and total energy intake per day (kcal/day).

        ‡Model 3 is adjusted for all factors in models 1 and 2 plus 24-hour urinary creatinine.

        §The signal at δ 6.91 of tyrosine also includes a signal from its metabolite 4-hydroxymandelate.

        Supplementary Materials

        • www.sciencetranslationalmedicine.org/cgi/content/full/7/285/285ra62/DC1

          Fig. S1. Distributions of the ICCs of the 1H NMR data.

          Fig. S2. Scatterplot of observed BMI (UK) versus BMI estimated using UK data by the U.S. EN model (tables S13 and S14) (F test: F52,391 = 4.04, P = 9.12 × 10–16).

          Table S1. Descriptive data of the INTERMAP study.

          Table S2. Chemical shifts identified and identification strategy for each 1H NMR metabolite.

          Table S3. ICC for each 1H NMR metabolite for the first and second urine collection data (obtained on average 3 weeks apart).

          Table S4. Structurally unidentified 1H NMR–derived signals associated significantly with BMI in 1880 U.S. INTERMAP participants using first urine collection specimens.

          Table S5. P value for a gender-interaction term included in the models for the U.S. INTERMAP population (n = 1880).

          Table S6. P value for a gender-interaction term included in the models for the UK INTERMAP population (n = 444).

          Table S7. Structurally unidentified 1H NMR–derived signals associated significantly with BMI in 1880 U.S. INTERMAP participants using second urine collection specimens.

          Table S8. Structurally identified 1H NMR–derived urinary metabolites associated significantly with BMI in 1880 U.S. INTERMAP participants using second urine collection specimens.

          Table S9. Association with BMI of metabolites identified by 1H NMR in 444 UK INTERMAP participants for second urine collection specimens.

          Table S10. Full names and abbreviations of metabolites in the metabolic reaction network (Fig. 3).

          Table S11. Abbreviations and closest affinity classifications of metabolites in the metabolic reaction network (Fig. 3).

          Table S12. Metabolite-BMI associations adjusted for all other metabolites.

          Table S13. EN model of all variables combined (λ = 0.15, α = 0.98).

          Table S14. EN model of all variables combined (λ = 0.17, α = 0.59), excluding population sample variables.

        • Supplementary Material for:

          Urinary metabolic signatures of human adiposity

          Paul Elliott,* Joram M. Posma, Queenie Chan, Isabel Garcia-Perez, Anisha Wijeyesekera, Magda Bictash, Timothy M. D. Ebbels, Hirotsugu Ueshima, Liancheng Zhao, Linda van Horn, Martha Daviglus, Jeremiah Stamler, Elaine Holmes, Jeremy K. Nicholson*

          *Corresponding author. E-mail: p.elliott{at}imperial.ac.uk (P.E.); j.nicholson{at}imperial.ac.uk (J.K.N.)

          Published 29 April 2015, Sci. Transl. Med. 7, 285ra62 (2015)
          DOI: 10.1126/scitranslmed.aaa5680

          This PDF file includes:

          • Fig. S1. Distributions of the ICCs of the 1H NMR data.
          • Fig. S2. Scatterplot of observed BMI (UK) versus BMI estimated using UK data by the U.S. EN model (tables S13 and S14) (F test: F52,391 = 4.04, P = 9.12 × 10−16).
          • Table S1. Descriptive data of the INTERMAP study.
          • Table S2. Chemical shifts identified and identification strategy for each 1H NMR metabolite.
          • Table S3. ICC for each 1H NMR metabolite for the first and second urine collection data (obtained on average 3 weeks apart).
          • Table S4. Structurally unidentified 1H NMR–derived signals associated significantly with BMI in 1880 U.S. INTERMAP participants using first urine collection specimens.
          • Table S5. P value for a gender-interaction term included in the models for the U.S. INTERMAP population (n = 1880).
          • Table S6. P value for a gender-interaction term included in the models for the UK INTERMAP population (n = 444).
          • Table S7. Structurally unidentified 1H NMR–derived signals associated significantly with BMI in 1880 U.S. INTERMAP participants using second urine collection specimens.
          • Table S8. Structurally identified 1H NMR–derived urinary metabolites associated significantly with BMI in 1880 U.S. INTERMAP participants using second urine collection specimens.
          • Table S9. Association with BMI of metabolites identified by 1H NMR in 444 UK INTERMAP participants for second urine collection specimens.
          • Table S10. Full names and abbreviations of metabolites in the metabolic reaction network (Fig. 3).
          • Table S11. Abbreviations and closest affinity classifications of metabolites in the metabolic reaction network (Fig. 3).
          • Table S12. Metabolite-BMI associations adjusted for all other metabolites.
          • Table S13. EN model of all variables combined (λ = 0.15, α = 0.98).
          • Table S14. EN model of all variables combined (λ = 0.17, α = 0.59), excluding population sample variables.

          [Download PDF]

        Stay Connected to Science Translational Medicine

        Navigate This Article