Gian-Marco Rignanese is Professor at the Ecole Polytechnique de Louvain (EPL) and Research Director at the F.R.S.-FNRS. He received his Engineering degree from the Université catholique de Louvain in 1994 and Ph.D. in Applied Sciences from the Université catholique de Louvain in 1998.
During his Ph.D., he also worked as a Software Development Consultant for the PATP (Parallel Application Technology Project), collaboration between CRAY RESEARCH and Ecole Polytechnique Fédérale de Lausanne (EPFL) in the group of Prof. Roberto Car. He carried his postdoctoral research at the University of California at Berkeley in the group of Prof. Steven Louie. In 2003, he obtained a permanent position at the Université catholique de Louvain.
In 2019, he was named APS Fellow for original efforts developing free license software in the field of electronic structure calculations, and high-throughput calculations in a broad range of materials types.
Gian-Marco Rignanese is especially active in high-throughput approaches collaborating to the Materials Project, computing:
Furthermore, he exploits machine learning to predict the properties of materials.
His group plays an active role in the development of different softwares (Abinit, Abipy, PyMatGen, FireWorks) and pseudopotentials through the PseudoDojo.
In the last decades, a number of materials databases have become available online (see e.g. Ref. [1] for an extensive, yet inevitably incomplete, list). In many cases, these can be accessed via a graphical web interface which targets a “low-throughput” human usage but that is not very well suited for a systematic “high-throughput” computational approach. In fact, in order to take full advantage of modern data‑analytics techniques, it is essential that these databases also become accessible through an application programming interface (API) as it is already the case for some of them [2-5]. It would actually be even more beneficial to have access to information originating from multiple databases as they often cover different material families and properties. Nonetheless, retrieving data from multiple databases is difficult since the available APIs are different from one database to another.
In order to overcome these problems, the OPTIMADE API was developed. It was designed so that it can be implemented without significant changes to the established back-end code, and, furthermore, adopting the API is straightforward for the end user. The OPTIMADE specification version 1.0.0 was released on 1 July 2020 [6]. It is supported by leading databases such as AFLOW, the Materials Cloud, the Materials Project, NOMAD, OQMD, ... Currently, the returned properties comprise both mandatory information about the structure (such as the elements the lattice vectors), as well as optional and database-specific information prefixed with the database name (e.g., _aflow_).
In this talk, I will outline some key features of the API specification. I will illustrate its usage through some examples. Finally, I will discuss how it would benefit from more fundamental work on an ontology for materials databases.
