Abstract
With ever more comprehensive data sets and complex analysis algorithms, reproducibility is becoming increasingly important to ensure trust in research results. Availability and stability of the employed software algorithms are, along with data availability, crucial aspects of reproducible research. Open and collaborative software development can increase stability and long term support of software as the development is not carried out by a single research group and hence maintenance is less dependent on individual developers or funding of a single research group. Heavily interdisciplinary sciences, such as for example metabolomics, also benefit from open collaborative software development, as researchers with different background will contribute functionality from their specific expert areas resulting thus in a potentially more complete software solution.
In this presentation Johannes Rainer will share his own experiences with reproducible research and open software development in metabolomics research. He will discuss advantages, but also potential pitfalls of open software development and highlight opportunities that openness and collaborative efforts bring to modern data science.