High-performance computing (HPC) is one of the pillars accelerating scientific discoveries. From finding relevant new drugs for diseases to exploring new materials, HPC is pushing the boundaries of many fields of science and engineering, but most prominently Chemistry, Physics, Energy, Materials Science, Nanotechnology and Biochemistry. The volume of information that’s generated daily, coming from the results of scientific calculations is increasing exponentially and researchers need to have tools to manage the data. But that’s no easy task, as the process involves different steps as capture, storage, search, share, analysis and visualization.
Spearheaded by ICIQ group leader Carles Bo, the institute’s Computational Chemistry groups set off to help solve this problem. Over five years ago, the researchers created in collaboration with Universitat Rovira i Virgili (URV), ioChem-BD a digital repository aimed to manage, store and publish Computational Chemistry Datasets. Since then, ioChem-BD has grown and adapted to become a useful tool for the community.
Now, the platform ioChem-BD has been included in Nature’s Scientific Data Recommended Data Repositories allowing the whole community to easily publish the results of computational chemistry research projects. With the aim of promoting wider data sharing and reuse, and to credit those that share, Nature Research (part of Springer Nature) created Scientific Data, a peer-reviewed, open-access journal for descriptions of scientifically valuable datasets and research that advances the sharing and reuse of scientific data.
Since its dawn, ioChem-BD has been an Open Data repository, a unique platform for computational chemistry – and related fields – results with a general approach. The platform allows researchers to further analyse data published by their peers, or start new studies based on previously existing work. Now, ioChem-BD is taking this one step beyond.
Open Source for Open Science
On its third revision the first one fully open source, the ioChem-BD platform has opened up its source code, allowing researchers worldwide to modify and implement new functionalities on the platform. “We wanted to involve the community in the development so that everyone can contribute,” explains Carles Bo, ioChem-BD project leader.
With some hundreds users, ioChem-BD not only allows the publication of the results of the numerical experiments, but also the parameters needed and the input files. “You can find the initial data someone else used so that you can reproduce the study. One of our main goals when setting the platform was to follow the FAIR principles that meant the data available on the platform had to be findable, accessible, interoperable and reusable, so available to any scientist to reproduce previous experiments,” explains Bo.
Following the same principle, ioChem-BD is hosted on the Barcelona Supercomputing Center (BSC) infrastructure thanks to a collaboration agreement signed in 2016. To Bo, “partnering with a distinguished institution such as the BSC provides the technical expertise that brings our platform to the forefront of the chemical data repository needs.”
Although the raw datasets generated by computational chemists as well materials scientists are diverse and will take time for the community to organize itself, ioChem-BD has proven to be a useful tool, standing up to the challenge. For instance, the International Journal of Quantum Chemistry highlighted the versatility of ioChem-BD to archive and store data generated with a variety of simulation packages in their submission guidelines. And more recently, on its third revision of the Open Research Data Management Plan, the European Research Council also included ioChem-BD as an example of repositories for open access Computational Chemistry research results.
If you wish to run your own instance of ioChem-BD, you may request your free copy here. If instead, you prefer to run an account in the servers at the Barcelona Supercomputing Center (BCS-CNS), you can apply for a free user account using this form.