EC Working Groups
CINECA WP4 coordinates 2 active open science working groups to enhance collaboration between EC projects:
Federated Analysis Working Group (FAWG)
Joint Synthetic Data Initiative (JSDI)
Federated Analysis Working Group (FAWG)
The FAWG brings together representatives from all of the EUCAN projects, and has expanded to include other major EC-funded projects such as ORCHESTRA, unCoVer, EJP-RD, B1MG and INTERVENE. Participation in the FAWG benefits all members by openly sharing the current state of play developments, exhibiting new tools, and enhancing opportunities for the development of cross project demonstrators in federated research and clinical applications between EC projects. Recent meetings have hosted presentations from a diverse set of speakers including:
Shadi Albarqouni on the application of federated learning to AI in medicine.
Patrick Ruch - on the Swiss Variant Interpretation Platform: a one-stop shop for clinical interpretation of variants in oncology https://www.expasy.org/resources/svip.
Christian Mertes on the German node of the fEGA
Upcoming talks for 2022-2023:
Sept 19, 13:00-14:00 UTC
Lauren Maxwell (Heidelberg) – “Implementation of FAIR across initiatives to share COVID-19 data and the relevance of semantic interoperability”Nov 21, 16:00-17:00 UTC
Melissa Cline (UCSC) BRCA Exchange - "Federated Analysis for Cancer Variant Interpretation"Jan 16th 2023, 14:00-15:00 UTC
Akis Linardos and Socayna Jouide El Kaderi (euCanSHare) – “Federated Learning in the Wild: A Multi Center Study in Breast Cancer Screening”March 20th 2023, 14:00-15:00 UTC
Prof Emily Jefferson (Dundee) - “Towards federated analysis on population level data”May 15th 2023, 13:00-14:00 UTC
Michelle Williams, Associate Director BHF Data Science centre – “Challenges and opportunities for federated learning in trusted research environments”
The FAWG meets bi-monthly. Please contact info@cineca-project.eu for more information.
Joint Synthetic Data Initiative (JSDI)
One of the main bottlenecks for developing federated analysis applications and technical demonstrators is the lack of freely available realistic test datasets. In year 2-3 of the project, CINECA developed a set of synthetic cohort datasets which were generated based on the phenotypic data from four of our participating cohorts - UK Biobank, CoLaus, H3Africa, and the CHILD Cohort Study. The JSDI is a collaboration between a subset of EUCAN projects, B1MG, and INTERVENE to work on the next generation of synthetic datasets, focussing on the development of algorithms for and the generation of synthetic genetic datasets. The working group is looking at
the technical challenges of generating scalable synthetic datasets for polygenic risk score applications.
continuing development of the CINECA synthetic cohort EUROPE UK1 dataset for the joint variant calling and eQTL analysis use cases.
Please contact info@cineca-project.eu for more information.