Drawing the Bow for Reproducibility.

Abstract

When Robin the Hood draws his bow (Fig. 1), one can always be sure that his arrows will hit the target with 100% accuracy and only marginal deviation. But what if the bowman is Little John, would he hit the target equally well? He may pull the bow differently and have a slight, uncorrected astigmatism. Or let's think of Robin using the bow of the Sheriff of Nottingham (don't ask how he got his hands on it); would he still reproduce his initial high performance? How about Little John with the bow of the Sheriff? What if it's not even a bow but a crossbow? So far, I didn't yet even touch the issue of using different arrows handcrafted by different manufacturers or the quality of the interpretation of the results by the referees! Questions over questions that cytometrists ask themselves and others almost every day. They may wish to use different instruments from different providers, different tags for target molecule identification, running the tests in different labs and using alternative computational tools. Who is hitting the target best and how can all be as good as Robin Hood? Many open questions. Consequently, every year reproducibility in cytometry experiments and cytometry data analysis is discussed in several publications and expert workshops like at the CYTO conferences 1, 2 and were the theme of several my editorials 3, 4. Thus, it was only logical to finally come up with a focused issue on Rigor and Reproducibility to address several of the points above. I have the pleasure to introduce you this February issue guest edited by Peter Lopez and Ruud Hulspas dedicated to Rigor and Reproducibility in cytometry experiments. I wish to thank both colleagues for picking this topic and providing this state-of-the-art focus on a really hot topic. Read their guest editorial (see page 105) where they will go more in depth than I want to do here. As Editor's Choice of the month, I selected the experimental work by Andrä and colleagues (see pages 171–183). Although it is not an integral part of the special issue, I think it fits well. Herein, the authors raising the question whether the physical stress of sorting living cells by a droplet sorter alters their physiological behavior. Read also the Commentary to this manuscript written by Sack and Bitar (this issue, page 168). I would also like to direct your attention to the manuscript by Bagwell and colleagues (this issue, page 184). It is not part of the special issue either, but it proposes a way to clean high-content mass cytometry data in an automated fashion form unwanted events that can lead to a higher level of reproducibility in data analysis.