Reproducible research in the study of biological coloration
The study of colour in nature has generated insights into fundamental evolutionary and ecological processes, and research into colour traits is a rapidly growing field (Kelber & Osorio, 2010). The ongoing interest in biological coloration has in part been driven by the increased availability of key technologies, including spectrometry and photography, and concurrent advances in methods for analysing colour data, such as visual models (e.g. Endler and Mielke, 2005, Kelber et al., 2003 and Stevens et al., 2007). While these developments are positive for the field, the increasingly complex analyses being run on ever greater amounts of data heighten the need for comprehensive methods reporting and diligent data management (Alsheikh-Ali et al., 2011 and Nekrutenko and Taylor, 2012). Our aim was to explore the state of reproducibility in the study of biological coloration, and to suggest simple ways in which it may be improved. We first outline common methods for studying biological coloration and present guidelines for comprehensive methods reporting. We then explore how well some of these important criteria have been reported in the literature. We also quantify the availability of publicly archived data and code and suggest some useful tools for increasing the reproducibility of colour trait research more broadly.
Please refer to publisher version or contact your library.