Three HetSys students spanning three different cohorts have contributed to a paper just published in the Journal of Open Source Science (JOSS) describing the matscipy package for materials science at the atomic scale in Python. Lakshmi Shenoy (Cohort 2) and Fraser Birks (Cohort 4) added new functionality to the code for modelling fracture in metals and ceramics, respectively, while Tom Rocke (Cohort 3) added tools to simulate dislocations and stacking faults. matscipy is developed jointly between James Kermode’s group in Warwick and Lars Pastewka’s group in Freiburg, and also receives contributions from the broader community. The paper overall includes 20 authors from 9 institutions.

JOSS is unusual in that it publishes papers describing software contributions rather than new research findings. It is a Diamond Open Access journal (articles are fully open access, without any publishing charges). The review process is carried out openly on GitHub and places a strong emphasis on good coding practices such as the inclusion of automated tests and documentation. This aligns well with HetSys’ training on sustainable research computing, ensuring that the software produced outlasts individual PhD projects and magnifiying its impact.

Lakshmi said “it was a great opportunity to be part of this project and to get some hands on experience in how software development works”. Fraser commented “it feels great to have been able to contribute to cutting edge scientific software from the very start of my PhD - I really have the Hetsys training to thank for that", and Tom added "the PX915 group project showed us the importance of the good coding practices that are required to develop a package like matscipy".