HBM 2013 Poster - Brainomics (hbm2013_ter.pdf)

Brainomics - A management system for exploring and merging heterogeneous brain mapping data

V. Michel , Y. Schwartz , P. Pinel , O. Cayrol , A. Moreno , JB. Poline , V. Frouin , D. Papadopoulos Orfanos

Logilab, Paris, France, INRIA Saclay Parietal Team, Paris, France,

Unicog, INSERM-CEA, Neurospin, Gif-sur-Yvette, France,4CEA, Neurospin, Gif-sur-Yvette, France Introduction http://www.brainomics.net/demo/ Number of large datasets for brain mapping have been released [1, 2].  Neuroimaging datasets more routinely include clinical data or genetics data.  Exploitation requires ➔ An efficient organization to integrate all the measures ➔ An easy access to the relevant information.  Neuroimaging [3] and genomics[4] databases are dedicated to their own field of research.

Brings together brain imaging and genetics data. Relies on a high-level query language (RQL). Solution based on CubicWeb, a semantic framework. Visualizing / exporting data in several formats. Query using RQL http://www.cubicweb.org/ ✔ Similar to the W3C's SPARQL [6]. ✔ Supports the basic operations (selection, insertion, etc.). ✔ Subquerying, ordering, couting, ... ✔ Data management framework, 10 years of industrial uses (e.g. [5]). ✔ Well established core technologies: SQL, Python, HTML5, Javascript. ✔ Licensed under LGPL since 2008. ✔ Used in production environments since 2005. ✔ Fine-grained security system coupled to the data model definition. ✔ Migration mechanisms controls model version / ensures data integrity. Data model ✔ Described in Python, using reusable modules called “cubes”. ✔ Modelisation of Scans, Questionnaires, Genomics results, Behavioral results, Subjects and Studies information. ✔ Data model optimized for large volumes (> 2000 subjects). ✔ Tested with several publicly available datasets [1, 2]. Conclusion Query all the female subjects of the database, with an age greater than 30 Any S WHERE S is Subject, S age > 30, S gender "female" Query all the Cmap scans of subjects with an age greater than 25, and that have a score greater than 4.0 for the "algebre" question Any SA WHERE S is Subject, S age > 25, X is QuestionnaireRun, X concerns S, A is Answer, A questionnaire_run X, A question Q, Q text "algebre", A value > 4, SA is Scan, SA concerns S, SA type "c map" Views ✔Each query result can be seen using different views. ✔HTML pages, ZIP files, spreadsheets, XCEDE XML, ... ✔May include processing (stat. maps computed on the fly). Contact: brainomics@logilab.fr ✔ Open source solution to manage brain imaging datasets and associated meta data. ✔ Powerful querying and reporting tool, customized for emerging imaging-genetics field. This work was supported by grants from the French National Reseach Agency (ANR GENIM; ANR-10-BLAN-0128) and (ANR IA BRAINOMICS; ANR-10-BINF-04). [1] http://openfmri.org/data-sets [2] fcon_1000.projects.nitrc.org/indi/abide/ [3] Olsen, M.D. (2007). The extensible neuroimaging archive toolkit. Neuroinformatics 5, 11–33 [4] Vallon-Christersson, J. (2009). ‘BASE - 2nd generation software for microarray data management and analysis’ [5] Publishing bibliographic records on the Web of data: opportunities for the BnF (French national Library). ESWC 2013 [6] http://www.w3.org/TR/rdf-sparql-query/