In my opinion, sharing research data is all about transparency, reproducibility, and - what might be even more important - efficiency. In my daily ife as a computational biophysicist, I perform lots of molecular dynamics simulations on proteins, which produce “clips” showing how the atoms in the molecular systems move over time. These trajectories store each atom’s position and velocity at discrete points in time and can reach sizes in the order of GB. As such simulations are computationally rather demanding, the field could benefit from RDM in the way of that simulations are not repeated unnecessarily by different researchers not knowing about the work and data produced by each other.

I always try to store my data in a structured way to facilitate its reproducibility in the future. However, I have not yet decided for a consistent “scheme”, in particular with respect to the simulations’ metadata, and my data is not publicly accessible.

In addition, we also do some method development in our group, which involves writing code. All our software is available on GitHub.