23.01.2015 Views

Landscapes Forest and Global Change - ESA - Escola Superior ...

Landscapes Forest and Global Change - ESA - Escola Superior ...

Landscapes Forest and Global Change - ESA - Escola Superior ...

SHOW MORE
SHOW LESS
  • No tags were found...

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

S. Joost et al. 2010. GEOME. Towards an integrated web-based l<strong>and</strong>scape genomics platform<br />

671<br />

development of GEOME is also a partner within the EU FP7 NEXTGEN, the first project in the<br />

area of conservation genetics that proposes a comparative analysis of whole genome data at the<br />

intraspecific level (http://nextgen.epfl.ch). NEXTGEN will also need GEOME’s expertise in the<br />

area of remote sensing, digital elevation models <strong>and</strong> other environmental data in general.<br />

5. Geo-environmental data in a HPC environment<br />

Environmental sciences are a data-intensive domain in which applications typically produce <strong>and</strong><br />

analyze a large amount of geospatial data. Moreover, due to the multi-disciplinary nature of<br />

environmental sciences (e.g. ecology, climate change, etc.), scientists need to integrate large<br />

amount of data distributed all around the world in different data centers. A local cluster is the<br />

first mean to upscale computational capacities when the workflow can be distributed into many<br />

independent jobs.<br />

When the number of jobs is very large <strong>and</strong>/or users do not belong to the Institution owning the<br />

cluster, grid computing can be an efficient solution. Following Foster et al. (2008), a grid is a<br />

parallel processing architecture in which computational resources are shared across a network<br />

allowing access to unused CPU <strong>and</strong> storage space to all participating machines. Resources could<br />

be allocated on dem<strong>and</strong> to consumers who wish to obtain computing power. Recent studies have<br />

had a successful approach to extend grid technology to the remote sensing community (Muresan<br />

et al. 2006), as well as in the field of disaster management (Mazzetti et al. 2009) making OGC<br />

web service grid-enabled.<br />

One of the largest scientific grid infrastructures currently in operation is the Enabling Grids for<br />

E-sciencE (EGEE) infrastructure bringing together more than 120 organisations to provide<br />

scientific computing resources to the European <strong>and</strong> global research community. The currently<br />

120'000 CPUs available in EGEE are essentially used by the High Energy physics community,<br />

but part of EGEE is also opened to other disciplines such as environmental sciences or genetics.<br />

A grid environment federates its users through Virtual Organizations (VOs), which are sets of<br />

individuals <strong>and</strong>/or institutions defined by a set of sharing rules (e.g. access to computers,<br />

software, data <strong>and</strong> other resources). One particular VO of interest is named Biomed; it has<br />

currently access to 20'000 CPUs <strong>and</strong> is willing to accept the envisioned GEOME on the Biomed<br />

VO.<br />

6. Conclusion<br />

No software tool presently exists to answer challenges of developing better approaches for<br />

linking ecologically relevant data sets to specific loci or genes. Moreover, no software tool can<br />

currently integrate geo-environmental variables <strong>and</strong> molecular data within a WebGIS<br />

environment, with analysis modules included i) to detect loci under selection according to two<br />

complementary approaches, ii) to analyze <strong>and</strong> characterize the spatial distribution of these loci,<br />

<strong>and</strong> iii) to produce predictive habitat maps derived from them.<br />

Thanks to a robust HPC infrastructure, GEOME’s existing <strong>and</strong> future Web services will be<br />

useful to a very large number of users, <strong>and</strong> will possibly contribute in underst<strong>and</strong>ing how<br />

l<strong>and</strong>scape-level geographical <strong>and</strong> environmental features are involved in the distribution of the<br />

functional adaptive genetic variation.<br />

References<br />

Bernard, L., <strong>and</strong> Craglia, M., 2005. SDI - From Spatial Data Infrastructure to Service Driven<br />

Infrastructure. First Research Workshop on Cross-learning on Spatial Data<br />

Infrastructures (SDI) <strong>and</strong> Information Incrastructures (II), Enschende, The Netherl<strong>and</strong>s.<br />

Blow, N., 2008. DNA sequencing: generation next-next. Nature Methods, 3: 267- +.<br />

<strong>Forest</strong> <strong>L<strong>and</strong>scapes</strong> <strong>and</strong> <strong>Global</strong> <strong>Change</strong>-New Frontiers in Management, Conservation <strong>and</strong> Restoration. Proceedings of the IUFRO L<strong>and</strong>scape Ecology<br />

Working Group International Conference, September 21-27, 2010, Bragança, Portugal. J.C. Azevedo, M. Feliciano, J. Castro & M.A. Pinto (eds.)<br />

2010, Instituto Politécnico de Bragança, Bragança, Portugal.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!