Getting the right answers: understanding metabolomics challenges

Abstract

Small molecules within biological systems provide powerful insights into the biological roles, processes and states of organisms. Metabolomics is the study of the concentrations, structures and interactions of these thousands of small molecules, collectively known as the metabolome. Metabolomics is at the interface between chemistry, biology, statistics and computer science, requiring multidisciplinary skillsets. This presents unique challenges for researchers to fully utilize the information produced and to capture its potential diagnostic power. A good understanding of study design, sample preparation, analysis methods and data analysis is essential to get the right answers for the right questions. We outline the current state of the art, benefits and challenges of metabolomics to create an understanding of metabolomics studies from the experimental design to data analysis.

Keywords:

• community standards
• data processing
• design of experiments
• mass spectrometry
• metabolites
• metabolome
• metabolomics
• nuclear magnetic resonance
• small molecules

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.

Key issues

• Metabolite identification beyond mere annotation. Achieving significant metabolite identifications without extensive in-house libraries and sample spiking is probably the most important issue in metabolomics.

• Community reporting and data exchange standards need to be adopted by the wider community to enable efficient data exchange and higher quality metabolomics research.

• ‘Wide matrices’, that is, matrices with a small number of samples but a large number of variables, generate spurious relationships that can be misleading. More is not always better. Better study designs and awareness of issues related to sample size need to penetrate the metabolomics community.

• Datasets generated from large-scale metabolomics experiments introduce great challenges in the context of random and systematic noise and cherry-picking of results. The application of robust statistical methods has never been more important.

• Collaborations between laboratories and researchers with different areas of expertise. Metabolomics is at the interface of many disciplines, and efficient and productive collaborations are challenging on multiple levels but of paramount importance.