Pages

Thursday, March 29, 2012

Matching the needs of users: a life sciences perspective

Whether it’s for a bioinformatics application or for visualising cells at a molecular level, it is clear life science communities are increasingly calling upon exascale computational capabilities, simulation and modelling applications and e-infrastructures to accelerate their research. Although there is enormous potential for integration and modelling on a multi-scale level, it is surprising that life scientists are still not major users of the grid (approximately 11% of life scientists use PRACE and 15% using national computing centres, and less are using the Grid itself). So it's great to see that many projects presented at the EGI Community Forum this week, are helping non-experts to adopt the technology.

On Wednesday morning, we heard from Alexander Bovin about the WeNMR project, who are successfully growing their life science user base (http://www.wenmr.eu/). WeNMR is the first and largest Virtual Research Community to be supported by EGI within the life science area. WeNMR provide simple user-friendly access to Grid via web portals. Two dozen computational NMR services are available on the platform, and a number of applications for modelling macromolecules e.g. 3D Dart, Haddock. Users have access to various training resources (including wikis, online tutorials, a blog, a help centre, NMR and SAXS services).


The average life science researchers often doesn't come from a computing background, and hence choosing the most appropriate application software/architecture to fit their problem can prove difficult.  The ScalaLife  project (presented by Professor Erwin Laure from the Royal Institute of Technology (KTH), Stockholm in Wednesday) is an ambitious initiative that hopes to provide a one-stop-shop knowledge-base and advice centre for the computational Life Science community. You can read more on the project in a blog entry yesterday from iSGTW journalist, Adrian Giordani. ScalaLife aims to help researchers by providing best practice, support and experience in order to help them develop more accurate structural and functional models, and in the longer-term could potentially have some impact in reducing the cost of generating new drugs.

Another project that resonated with me was ScienceSoft (http://sciencesoft.web.cern.ch/), which again is a new initiative to assist scientific communities in finding the software they need. The project is currently in a early scoping phase, but hopes to promote the development and use of open source software for scientific research. The project will provide a marketplace to match user needs to software products and services. Possible content could include catalogues, software services, a skills database, information on who likes what and why (a rating system), a peer review system as well as a citation system to allow software to be references in papers. At the moment, ScienceSoft is looking for ideas from community members, editors and developers to offer advice on where functionality is needed. GridCast interviewed Alberto Di Meglio about the initiative in a video featured yesterday afternoon. 

No comments: