To the editor:

We read with great interest the commentary by John Lindon et al. in the July issue (Nat. Biotechnol. 23, 833–838, 2005) describing the Standard Metabolic Reporting Structures (SMRS) community-driven initiative to establish standards for the reporting of metabolic analyses. This initiative follows on from successful initiatives in other disciplines, such as functional genomics (Microarray Gene Expression Data; MGED)1 and proteomics (Human Proteome Organization Proteomics Standardization Initiative; HUPO-PSI)2, and is an important first step toward permitting the sharing of high-quality, structured data both within the discipline of metabolomics and with other disciplines, although we acknowledge that developing standards for metabolic data may be more complex as the data are more diverse.

The development of standards has been shown to have huge economic benefits3 and is a fundamental requirement to enable data sharing, as implemented by the US National Institutes of Health (NIH) and described by the UK National Cancer Research Institute (NCRI)4. We particularly advocate the approach of early-stage community involvement in the development of such standards as this is the most effective way to ensure that the output generated is both pragmatic and useful, and that such standards are accepted by the community.

This is a major purpose of the NCRI Informatics Initiative supported by the 20 major UK cancer funders in government, academia and industry. The NCRI Informatics Initiative is taking a broader view than the initiatives mentioned above in that it aims to integrate standards, data and resources from across the spectrum of cancer-related science and medicine and has produced a platform reference model that will guide this integration5. A preliminary resource-mapping exercise has been completed6 and we invite the community to use and contribute to development of the content of the matrix, which now refers to the SMRS working group, and welcome such initiatives in other disciplines.