Implementing code review in the scientific workflow: Insights from ecology and evolutionary biology

Abstract

Abstract Code review increases reliability and improves reproducibility of research. As such, code review is an inevitable step in software development and is common in fields such as computer science. However, despite its importance, code review is noticeably lacking in ecology and evolutionary biology. This is problematic as it facilitates the propagation of coding errors and a reduction in reproducibility and reliability of published results. To address this, we provide a detailed commentary on how to effectively review code, how to set up your project to enable this form of review and detail its possible implementation at several stages throughout the research process. This guide serves as a primer for code review, and adoption of the principles and advice here will go a long way in promoting more open, reliable, and transparent ecology and evolutionary biology. Abstract Code review is the process of either informally (as part of a group, as colleagues) or formally (as part of the peer review process) checking and evaluating each other's code and is a critical method of reducing errors and increasing research reproducibility and reliability. In this paper, we provide a detailed commentary on how to effectively review code (including introducing the four Rs), how to set up your project to enable this form of review and detail its possibleimplementation at several stages throughout the research process.