Case Reports of Pre-clinical Replication Studies in Metabolism and Diabetes.

Recent articles have highlighted the lack of reproducibility of data from scientific publications. Here we would argue that a better way to describe and also tackle this matter is to use the term "lack of robustness," since it points toward potential solutions. Presenting several case reports, we highlight examples with common underlying issues from Novo Nordisk's experience: animal model variability, reagent quality, and inter-lab variability. We discuss means to prevent these issues and argue for increased collaborative work and transparent manuscript revision procedures. Collectively, we believe these measures will help promote a more rapid and efficient self-corrective process in diabetes drug target research.