Have you ever felt you could solve a problem in your research if you could just visualise it as a model, analyse your data with an algorithm made just for you and think about it from another disciplinary perspective ?
The institute for computing and data science (ISCD) was founded for this reason: we are dedicated to exploring, developing and promoting innovative mathematical and computational research across sciences, humanities and medicine.
Using our unique expertise in scientific computing and data analysis, we bring together scientists from different disciplines so that they can work together on a research project, while remaining active members of their home faculties.
If you want to go far, go together
It all started when a group of researchers at UPMC realized that they were facing similar challenges because they were all using high-performance computing. Instead of trying to solve these problems in their separate labs, they decided to come together to discuss interdisciplinary research with HPC resources.
What began as an informal work group soon became an institute where researchers from different disciplines could come together to share their expertise in order to solve complex problems with the help of large scale computation, data analysis and scientific visualisation. Research in fields like chemistry, physics, biology, medicine and the humanities led in turn to new developments in mathematics and informatics.
The institute is determined to foster this type of research because we know it will become increasingly common in the comming years as we try to solve multidimensional problems.
As the proverb says: if you want to go fast, go alone. If you want to go far, go together.
Computing + Data = Knowledge
Finding the information you want in a large quantity of data often feels like looking for a needle in a haystack. Sometimes this is because of the sheer quantity of data that your algorithm has generated – the human brain just can’t process it. And when you want to analyse your data, you can’t always be sure that your original data will meet your needs. So, in many cases, data is made into knowledge through educated guesswork.
Part of the problem is that we’ve separated the making of collecting data from data analysis: we either generate data from computing or collect and analyse data with algorithms. What we try to do at the institute is to combine both computing and analysis as much as we can. We develop tailor-made software that provides our researchers with data that is both precise and suited to their research. The way we proceed varies widely from project to project and we still work with data collected in the field, but always with an eye toward the quality of the data.
This allows us to reconstruct faces from skulls, rebuild ancient architecture, predict health risks, understand protein interactions, program supercomputers, etc.
The annual budget of the ISCD is provided by Sorbonne Université. In 2012, the ISCD has successfully applied to the French investment program (PIA/Programme Investissements d’Avenir) which acknowledged the structuring nature of this project to tackle interdisciplinary challenges in scientific computing. Since then, additional fundings came through European, national and regional initiatives. Sponsorship provides opportunities to have access to a valuable resource that would be difficult to be accommodated by the core institutions.
A measure of the success of the ISCD model is the amount of third-party funding raised by its researchers. A selection of major grants obtained by the researchers shows that the ISCD has been able to raise substantial funding to sustain its activities.
- 2011-2019 equip@meso, tremplin vers le calcul pétaflopique, Programme Investissements d’Avenir, 1.3M€
- 2011-2017 Sesame RefICS, Plateforme Paris Ile de France pour la Recherche et la Formation interdisciplinaires en calcul et simulation, 720k€
- 2012-2019 Calsimlab, Labex pour la modélisation et la simulation scientifiques en recherche, Programme Investissements d’Avenir, 5M€
- 2014-2018 Chair FaciLe, Reconstruction cranio-faciale numérique pour la médecine légale, 1.2M€
- 2015-2018 ICS@SU, Action structurante: institut des sciences du calcul et des données, 720k€
- 2015 FEDER, Fédération calcul scientifique, analyse et visualisation des données complexes, 380k€