SKM 2023 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
SOE: Fachverband Physik sozio-ökonomischer Systeme
SOE 13: Data Analytics of Complex Dynamical Systems (joint session DY/SOE)
SOE 13.4: Talk
Thursday, March 30, 2023, 10:15–10:30, MOL 213
Reproducibility of analysis workflows in biomedical physics — •Alexander Schlemmer1,4, Inga Kottlarz1,2, Baltasar Rüchardt1,3,4, Ulrich Parlitz1,2,4, and Stefan Luther1,2,3,4 — 1Max Planck Institute for Dynamics and Self-Organization, Göttingen, Germany — 2Institute for the Dynamics of Complex Systems, Georg-August-Universität Göttingen, Germany — 3Institute of Pharmacology and Toxicology, University Medical Center Göttingen, Germany — 4German Center for Cardiovascular Research (DZHK), Partner Site Göttingen, Germany
Sustainable and well-documented data analysis workflows are essential for effectiveness and reproducibility in data-intensive research. In our terminology, documentation includes method and algorithm descriptions as well as human- and machine-readable representations of parameters, initial conditions and data, versions and dependencies and a well-defined software execution environment.
In practice, many software frameworks for reproducibility fail to achieve a widespread adoption. Using examples from data analysis in cardiac research, we illustrate typical challenges and show, how simple guidelines - when implemented in a pragmatic way - can already lead to a high degree of documentation and reproducibility. Furthermore, we discuss the employment of containers and semantic data management which simplify reproducibility, findability and interoperability.