SURFsara

The capability SURFsara is unique in the Netherlands as we operate at the front-end of innovative solutions, creating state-of the art technology. SURFsara provides the national e-infrastructure for research institutes in the Netherlands.

Genome analysis and imputation, BBMRI consortium: This research focuses on the analysis of 750 human genomes, connecting genotype changes to differences in phenotype. Estimating genetic variants based on data collected in genome-wide association studies combined with whole genome NGS datasets were run on the grid infrastructure, the life science grid. These applications are very data and computationally intense. They use a pilot job framework developed within the group itself called RITE. SURFsara’s hardware resources and expertise played an important role in building the e-infrastructure for this BBMRI subproject.

Medical imaging, AMC: The promise of extracting objective information to characterize disease (e.g. biomarkers of disease) even before it becomes symptomatic and using this information to produce patient-specific treatment (drug development) makes modern biomedical research a scientific field that is both data-intensive (vast amounts of data, heterogenous, distributed) and compute-intensive (sophisticated  analysis and simulation methods). The Life Science Grid has offered an infrastructure for scientists within the BioMed virtual Grid organisation to target these challenges.

Comparative proteomics, TU Delft: This research involves pattern recognition and machine learning in establishing and modelling protein networks. Typically, running a number of algorithms that need to be run across a variety of datasets, requires many millions of jobs which all have a limited input set and very small outputs (in the KB range). These outputs are saved to the pilot job system itself (in this case PiCaS) as this data cannot efficiently be stored on a SRM. Run times for these applications are in the order of a few hundred thousand core hours per experiment. SURFsara provided its expertise and hardware resources to make this project to a success.

Systems biology. NCSB, CWI: This Netherlands Consortium for Systems Biology (NCSB) project requires an upscale of simulations of populations of metabolizing bacteria, that are used to study behavior and evolution of gut microbiota. In silico simulation models of the gut microbiota aim to uncover fundamental principles of intestinal microbial organization and its relation to health and diet of the host. We will use this VO playground project to set up a WS-VLAM infrastructure supporting our simulation system and to perform some initial simulations with it. As a first goal we aim at identifying parameter regions that support the coexistence of multiple bacterial species. This project is in a pilot stage in order to collaboratively set up an e-infrastructure for this life science research use case.

n/a

The services are offered under the “flag” of e-Infra (SURFsara, Groningen University,
and Nikhef), it is run by SURFsara. SURFsara is part of SURF, the collaborative organisation for ICT in Dutch education and research.

SURFsara works closely together with NBIC, NLeSC and the life science community
including UMCs, and is involved in a range of national and international
projects such as BBMRI and TraIT. As a member of the EGI ELIXIR virtual team,
as well as being partner in several EU projects, we share our expertise to
build EU federated e-infrastructure solutions.