Taking On The Problem Of Reproducibility

The hunt for biomarkers—measurable indicators that provide information about a biological or clinical state—has been plagued by problems with reproducibility. Many studies that identify potential biomarkers can’t be replicated by other labs, resulting in years of wasted effort and resources. Two organizations are confronting the problem head-on, with an emphasis on standards.

In December, the Global Biological Standards Institute released “The Case for Standards.” In this report, it argued that most irreproducibility in life sciences research “can be traced to the lack of a unifying standards framework.” What scientists need, for example, are highly characterized reference materials and document-based consensus standards on the conduct and reporting of research.

And last week, the National Biomarker Development Alliance (NBDA), a nonprofit sponsored by the Aerospace & Defense Research Collaboratory at Arizona State University, made its debut at a press conference in Washington, D.C. At the launch, Anna D. Barker, president and director of NBDA, announced plans to develop biomarker standards and to help validate biomarkers reported in the literature.

Journals can help improve reproducibility. In her Jan. 17 editorial, Marcia McNutt, editor-in-chief of Science, announced that the publication is adopting U.S. National Institute of Neurological Disorders & Stroke recommendations that preclinical studies be designed and reported in ways that make it easier for other labs to replicate the findings.