CORE: Comparable, Open, Reliable, Extensible Software: An Experience Report of Four Flux Balance Analysis Tools

Computational tools have become increasingly important to the advancement of biological research. Despite the existence of datasharing principles such as FAIR, there has been little attention paid to ensuring that computational tools provide a platform to compare experimental results and data across different studies. Tools often are lightly documented, non-extensible, and provide different levels of accuracy of their representation. This lack of standardization in biological research reduces the potential power for new insight and discovery and makes it hard for biologists to experiment, compare, and trust results between different studies. In this poster we present our experience using four tools that perform flux balance analysis, on a set of different metabolic models. We frame our work around a proposed principle, akin to FAIR, aimed at bioinformatics tools. We call this CORE (Comparable, Open, Reliable and Extensible), and find the biggest challenges to be with comparability between tools. We also find that while the tools are all open source, without a deep understanding of the code base, they have insufficient openness. We needed to reach out to developers to resolve many of our questions, and we were still left with unexpected behaviors. We present lessons learned as a path to future improvement using the CORE principles.