Improving reproducibility in animal research by splitting the study population into several 'mini-experiments'

Area of Science:

• Life Sciences
• Animal Research
• Experimental Design

Background:

• The scientific community faces a 'reproducibility crisis', impacting research reliability.
• Multi-laboratory studies improve external validity but are resource-intensive.
• A need exists for single-laboratory strategies to enhance reproducibility.

Purpose of the Study:

• To propose and validate a new experimental strategy for improving reproducibility within a single laboratory.
• To adapt the multi-laboratory design logic to a single-laboratory setting.
• To empirically test a 'mini-experiment' design against conventional standardized approaches.

Main Methods:

• Systematically introduced heterogeneity by dividing experiments into 'mini-experiments' conducted over time.
• Compared a 'mini-experiment' design with a conventionally standardized design across independent replicates.
• Investigated behavioural and physiological differences between four mouse strains.

Main Results:

• The 'mini-experiment' design significantly improved reproducibility in approximately half of the investigated mouse strain comparisons.
• This strategy enhanced the accurate detection of treatment effects.
• The proposed design proved effective in tackling poor reproducibility in single-laboratory studies.

Conclusions:

• A practical and validated strategy for enhancing reproducibility in single-laboratory studies has been developed.
• The 'mini-experiment' approach offers a viable solution to the reproducibility crisis in life sciences.
• This method is adaptable to various experiments sharing similar characteristics across disciplines.