Advertisement

Advanced Statistical Methods for NMR-Based Metabolomics

Protocol
First Online:
Part of the Methods in Molecular Biology book series (MIMB, volume 2037)

Abstract

Despite the increasing popularity and applicability of metabolomics for putative biomarker identification, analysis of the data is challenged by low statistical power resulting from the small sample sizes and large numbers of metabolites and other omics information, as well as confounding demographic and clinical variables. To enhance the statistical power and improve reproducibility of the identified metabolite-based biomarkers, we advocate the use of advanced statistical methods that can simultaneously evaluate the relationship between a group of metabolites and various types of variables including other omics profiles, demographic and clinical data, as well as the complex interactions between them. Accordingly, in this chapter, we describe the method of seemingly unrelated regression that can simultaneously analyze multiple metabolites while controlling the confounding effects of demographic and clinical variables (such as gender, age, BMI, smoking status). We also introduce penalized orthogonal components regression as a screening approach that can handle millions of omics predictors in the model.

Key words

Seemingly unrelated regression Penalized orthogonal components regression Supervised dimension reduction Omics data 
This is a preview of subscription content, log in to check access.

Notes

Acknowledgments

This research was partially supported by NIH (R03CA211831, P30CA015704-40) and NSF CAREER award IIS-0844945. The authors gratefully acknowledge the support of the Cancer Care Engineering (CCE) project, a joint effort between the Oncological Sciences Center (Purdue Center for Cancer Research, NCI P30CA23168) in Purdue University Discovery Park and the Indiana University Melvin and Bren Simon Cancer Center (NCI P30CA082709).

References

  1. 1.
    Sud M, Fahy E, Cotter D, Azam K, Vadivelu I, Burant C et al (2016) Metabolomics workbench: an international repository for metabolomics data and metadata, metabolite standards, protocols, tutorials and training, and analysis tools. Nucleic Acids Res 44(D1):D463–D470CrossRefGoogle Scholar
  2. 2.
    Kraus WE, Muoio DM, Stevens R, Craig D, Bain JR, Grass E et al (2015) Metabolomic quantitative trait loci (mQTL) mapping implicates the ubiquitin proteasome system in cardiovascular disease pathogenesis. PLoS Genet 11(11):e1005553.  https://doi.org/10.1371/journal.pgen.1005553CrossRefPubMedPubMedCentralGoogle Scholar
  3. 3.
    Blasco H, Nadal-Desbarats L, Pradat PF, Gordon PH, Madji Hounoum B, Patin F et al (2016) Biomarkers in amyotrophic lateral sclerosis: combining metabolomic and clinical parameters to define disease progression. Eur J Neurol 23(2):346–353CrossRefGoogle Scholar
  4. 4.
    Nagana Gowda GA, Djukovic D (2014) Overview of mass spectrometry-based metabolomics: opportunities and challenges. Methods Mol Biol 1198:3–12CrossRefGoogle Scholar
  5. 5.
    Gromski PS, Muhamadali H, Ellis DI, Xu Y, Correa E, Turner ML et al (2015) A tutorial review: metabolomics and partial least squares-discriminant analysis–a marriage of convenience or a shotgun wedding. Anal Chim Acta 879:10–23CrossRefGoogle Scholar
  6. 6.
    Alakwaa FM, Chaudhary K, Garmire LX (2017) Deep learning accurately predicts estrogen receptor status in breast cancer metabolomics data. J Proteome Res 17:337–347CrossRefGoogle Scholar
  7. 7.
    Cuperlovic-Culf M (2018) Machine learning methods for analysis of metabolic data and metabolic pathway modeling. Metabolites 8(1): pii: E4. doi:  https://doi.org/10.3390/metabo8010004CrossRefGoogle Scholar
  8. 8.
    Chen C, Deng L, Wei S, Nagana Gowda GA, Gu H, Chiorean EG et al (2015) Exploring metabolic profile differences between colorectal polyp patients and controls using seemingly unrelated regression. J Proteome Res 14(6):2492–2499CrossRefGoogle Scholar
  9. 9.
    Chen C, Nagana Gowda GA, Zhu J, Deng L, Gu H, Chiorean G et al (2017) Altered metabolite levels and correlations in patients with colorectal cancer and polyps detected using seemingly unrelated regression analysis. Metabolomics 13:125.  https://doi.org/10.1007/s11306-017-1265-0CrossRefPubMedPubMedCentralGoogle Scholar
  10. 10.
    Zhang D, Lin Y, Zhang M (2009) Penalized orthogonal-components regression for large p small n data. Electronic J Stat 3:781–796CrossRefGoogle Scholar
  11. 11.
    Zellner A (1962) An efficient method of estimating seemingly unrelated regressions and tests for aggregation bias. J Am Stat Assn 57(298):348–368CrossRefGoogle Scholar
  12. 12.
    Lin Y, Zhang M, Wang L, Pungpapong V, Fleet JC, Zhang D (2009) Simultaneous genome-wide association studies of anti-CCP in rheumatoid arthritis using penalized orthogonal-components regression. BMC Proc (Suppl 7):S20CrossRefGoogle Scholar
  13. 13.
    Lin Y, Zhang M, Zhang D (2015) Generalized orthogonal-components regression for high-dimensional generalized linear models. Comput Stat Data Anal 88:119–127CrossRefGoogle Scholar
  14. 14.
    Wang L, Pungpapong V, Lin Y, Zhang M, Zhang D (2011) Genome-wide case-control study in GAW17 using coalesced rare variants. BMC Proc 5(Suppl 9):S110CrossRefGoogle Scholar
  15. 15.
    Zhang M, Lin Y, Wang L, Pungpapong V, Fleet JC, Zhang D (2009) Case-control genome-wide association studies of rheumatoid arthritis from GAW16 using POCRE-LDA. BMC Proc Suppl 7:S17CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2019

Authors and Affiliations

  1. 1.Department of StatisticsPurdue UniversityWest LafayetteUSA

Personalised recommendations

Cite protocol

Buy options