Guest post by Mark Viant, Professor of Metabolomics in the School of Biosciences at the University of Birmingham, UK, and Director of both the national NERC Biomolecular Analysis Facility – Metabolomics and the Phenome Centre Birmingham.
In 2014, my research team published the first Scientific Data Data Descriptor for metabolomics measurements, Direct infusion mass spectrometry metabolomics dataset: a benchmark for data processing and quality control. This article described in great detail the many steps that are critical for ensuring the production of high quality (direct infusion) mass spectrometry (DIMS) data. It was our intention that this publication would help to establish the benchmark for DIMS metabolomics, derived using best-practice workflows and rigorous quality assessment. The data was also made freely available in the MetaboLights public database for metabolomics data (dataset MTBLS79).1
Approaching two years later, I now ask to what extent the Scientific Data publication achieved our goals, and whether the reporting and standardisation of data within the metabolomics community has changed during this period, if at all.
Reviewing the citations for our Data Descriptor it is clear that it’s mainly being utilised as a methodological benchmark. Not only does this reassure me that the investment of my teams’ time in writing the manuscript was justified, but also highlights a more general point about the value of publishing data descriptors of the type reported in Scientific Data. Specifically, this publication has not only provided our dataset with greater visibility than would occur through an original research article, but importantly has raised the profile of the analytical and computational methods that we used for collecting and analysing the data. Such a publication channel, subject to robust reviewing, should have a major beneficial impact on data quality in the life sciences.
Second, I wish to comment on the importance of international data repositories. In metabolomics there are primarily two international data repositories, the MetaboLights repository based at EMBL-EBI in Europe1 and Metabolomics Workbench sponsored by the US National Institutes of Health and housed at UC San Diego.2 To what extent though is the metabolomics community using these resources, and is their use increasing?
In October 2014 we completed an international survey that was designed to determine the training needs of the metabolomics community, organised on behalf of the international Metabolomics Society and ELIXIR-UK. That survey revealed a rather disappointing observation, that of the 202 respondees only ca. 22% had used MetaboLights and ca. 12% had used Metabolomics Workbench.3 In February 2016 we completed a further international survey, this time focused on computational workflows, and again asked about the use of repositories. I am happy to report that the use of data repositories is increasing; specifically the survey revealed that 31% of scientists are now using MetaboLights and 14% using Metabolomics Workbench (the results of this survey are currently unpublished). This finding is consistent with the observation that MetaboLights is currently the fastest growing repository at EMBL-EBI (from personal communication with Christoph Steinbeck, Team Leader in Cheminformatics and Metabolism at EBI).
I believe the intrinsic value of international data repositories is being recognised at all levels – by both ‘wet’ and ‘dry’ lab researchers. This is in part to ensure that the appropriate reference data are available to maximally leverage discoveries from new data. I hope this increasing trend continues and more than 50% of the community are utilising these incredible resources by the end of 2016.
A further topic of great importance, and one that has been somewhat neglected in metabolomics, concerns reporting standards. In 2007, the significant activity of the international Metabolomics Standards Initiative yielded a journal full of manuscripts describing recommended reporting standards for a wide range of metabolomics studies, spanning the biological descriptors, analytical methods and data analysis approaches4 and further references in that edition of the Metabolomics journal]. Some of those efforts have been well cited, for example the Proposed minimum reporting standards for chemical analysis Chemical Analysis Working Group (CAWG) Metabolomics Standards Initiative (MSI)5 has 603 citations, according to Google Scholar. Yet there has been no sustained effort at refining and updating these early recommendations, something which is sorely needed by the metabolomics community.
New efforts are emerging, for example the Metabolomics Society has two task groups that are working in this area: the Data Standards task group, with the aim to foster and coordinate efforts in enabling efficient storage, compression, exchange and verification of information within metabolomics datasets; and the Metabolite Identification task group that includes within its aims the goal of building consensus on metabolite identification reporting standards. Again, journals such as Scientific Data can also play an important role in raising the profile of high quality and rigorous reporting standards.
I strongly encourage the metabolomics community to engage in this important process and work towards an improved set of reporting requirements. This activity should be encouraged through journal publishers requiring that minimal information standards for a metabolomics experiment are met to ensure compliance with open and transparent publishing.
- Haug, K.; Salek, R.M.; Conesa, P.; Hastings, J.; de Matos, P.; Rijnbeek, M.; Mahendraker, T.; Williams, M.; Neumann, S.; Rocca-Serra, P.; et al.
MetaboLights– an open-access general-purpose repository for metabolomics studies and associated meta-data
Nucl. Acids Res. 2013, 41, D781–D786. doi: 10.1093/nar/gks1004
- Sud, M.; Fahy, E.; Cotter, D.; Azam, K.; Vadivelu, I.; Burant, C.; Edison, A.; Fiehn, O.; Higashi, R.; Nair, K.S.; et al.
Metabolomics Workbench: An international repository for metabolomics data and metadata, metabolite standards, protocols, tutorials and training, and analysis tools
Nucl. Acids Res. 2016, 44, D463–470. doi: 10.1093/nar/gkv1042
- Weber, R.J.M.; Winder, C.L.; Larcombe, L.D.; Dunn, W.B.; Viant, M.R.
Training needs in metabolomics
Metabolomics 2015, 11, 784–786. doi: 10.1007/s11306-015-0815-6
- Fiehn, O.; Robertson, D.; Griffin, J.; van der Werf, M.; Nikolau, B.; Morrison, N.; Sumner, L.W.; Goodacre, R.; Hardy, N.W.; Taylor, C.; Fostel, J.; Kristal, B.; Kaddurah-Daouk, R.; Mendes, P.; van Ommen, B.; Lindon, J.C.; Sansone, S.-A.
The metabolomics standards initiative (MSI)
Metabolomics 2007, 3, 175-178.
- Sumner, L.W.; Amberg, A.; Barrett, D.; Beale, M.H.; Beger, R.; Daykin, C.A.; Fan, T.W.-M.; Fiehn, O.; Goodacre, R.; Griffin, J.L.; Hankemeier, T.; Hardy, N.; Harnly, J.; Higashi, R.; Kopka, J.; Lane, A.N.; Lindon, J.C.; Marriott, P.; Nicholls, A.W.; Reily, M.D.; Thaden, J.J.; Viant, M.R.
Proposed minimum reporting standards for chemical analysis Chemical Analysis Working Group (CAWG) Metabolomics Standards Initiative (MSI)
Metabolomics 2007, 3, 211-221.