An algorithm to match animals and increase statistical power and better distinguish treatment effect

Published 18 May 2017

In the latest Sciences Translational Medicine editorial, Aittokallio, Scherer, Poutanen and Freedman describe a mathematical algorithm that matches animals based on all of their baseline variables and environmental factors. The matching tool – freely available through a Web-based user interface – allows the allocation of animals to balanced intervention groups and the reduction of numbers of animals required to detect true treatment effects. In addition, the model-based power analysis informs the researcher if the planned sample size is not sufficient for detecting treatment effects avoiding underpowered studies.

The online tool is a timely effort towards reducing the reproducibility crisis which has been dramatically impacting the foundation of biomedical research in the past few years. The preclinical research community is invited to try out this new tool and provide input and suggestions on how to further improve its performance: 

The full article is available online

About the authors:

Tero Aittokallio is an EMBL Group Leader at the Institute for Molecular Medicine Finland (FIMM)

Andreas Scherer is a Project Director at FIMM, and National Coordinator in Finland for the European Infrastructure for Translational Medicine (EATRIS)

Matti Poutanen is a Professor in the Department of Physiology and Director of Turku Center for Disease Modeling (TCDM), Finland

Leonard P. Freedman is the Founding President of the Global Biological Standards Institute (GBSI), Washington