Diagnosing genetic diseases presents significant challenges. Modern sequencers produce vast amounts of data, requiring intensive processing and filtering before analysis. The intricate mechanisms in rare diseases necessitate a high level of expertise from biologists to identify potential causes. This is why it is crucial to demonstrate these results come from accurate, efficient, and reproducible bioinformatics pipelines. Indeed, accuracy is vital for patient diagnosis, efficiency is needed to handle the growing data volume and patient backlog, and reproducibility is essential to ensure the quality of the results.

This talk is aimed at a diverse audience and will give feedback on Nix integration in our workflow. It will focus on how Nix significantly improved reproducibility and discuss some challenges we encountered, especially with large databases and running on high-performance computing architectures.