TADA! Simple guidelines to improve analytical code sharing for transparency and reproducibility

Edward Richard Ivimey-Cook , Antica Culina, Shreya Dimri , Matthew Grainger, Fonti Kar, Malgorzata Lagisz , Nicholas Patrick Moran , Shinichi Nakagawa , Dominique G. Roche, Sean Tattan, Alfredo Sanchez-Tojar, Saras M Windecker, Joel L Pick 

Code sharing is essential to ensure transparency and computational reproducibility of published research, which in turn increases trust in scientific results. However, despite the growing number of journals that mandate code sharing, the prevalence of open code remains low, and substantially lags behind that of open data. Furthermore, even when it is openly shared, code is often non-functional, which hinders computational reproducibility. One reason for low levels of code sharing is uncertainty around how to properly archive functional analytical code associated with published research. Existing resources for best coding practices often do not sufficiently address how to archive analytical code, do not adhere to the established FAIR (Findable, Accessible, Interoperable, Reusable) principles, or are complex and primarily developed for software. To address this gap, we provide simple code sharing guidelines: TADA (Transferable, Available, Documented and Annotated). TADA details the minimum requirements necessary for a researcher to produce functional code for sharing that directly supports best practices and complements the FAIR principles. TADA aims to streamline the process of archiving and sharing functional code for researchers across all levels of coding experience, with the goal of increasing transparency, reproducibility, and the reliability of research results. Although these guidelines were developed based on our experience in Ecology and Evolutionary Biology, we believe they will be useful to researchers in other disciplines.