Never replicate a successful experiment? Standardization, heterogenization and reproducibility in animal experiments

点击翻译按钮获取中文摘要

Reproducibility is a cornerstone of any scientific method, especially in animal research where the lives of the animals are highly valuable. Traditionally, animals, husbandry and test procedures are standardized in animal experimentation to maximize test sensitivity and minimize animal use, assuming that this will also guarantee reproducibility. However, by reducing within-experiment variation, standardization may limit inference to the specific experimental conditions. Indeed, we have recently shown in mice that standardization may generate spurious results in behavioural tests, accounting for poor reproducibility. However, despite the widespread recognition of these problems, it is still not clear how to conduct an experiment within a single lab that may result in more stable results across studies. With the aim of making study populations more representative, systematic variation within studies (i.e. heterogenization) rather than rigorous standardization has been proposed to be a powerful tool to improve the reproducibility and the translational value of research findings from animal experiments Thus, the overall aim of the present project is to develop and validate practicable and effective heterogenization strategies that may serve to improve the reproducibility and translational value of research findings from animal experiments. In a series of pre-experiments, three different heterogenization strategies will be investigated and compared with respect to their effectiveness before testing the best strategy in a real multi-laboratory-setting. Factors that will be used for systematic heterogenization are (i) batch, (ii) day time, and (iii) experimenter, since all three factors are known to greatly influence the outcome of an experiment. To mimic a genotype effect, three commonly used inbred strains of mice (C57BL/6, DBA/2, Balb/c) will be examined in a battery of six different behavioural tests. Since limited reproducibility and translational failures has been particularly discussed in the context of behavioural phenotyping studies, we will restrict our analysis to behavioural outcome measures. Based on previous findings, we hypothesize that heterogenized experiments will yield more stable results across experiments than standardized ones.