dexterpratt
commented
5 years ago We would love to have NDEx (www.ndexbio.org) provide the data store for the application. NDEx is already tightly integrated with Cytoscape desktop and is a core technology of the emerging Cytoscape Web. REST transactions can be performed via CyREST or from a web application. There is an existing javascript library for translation of Cytoscape desktop graphic styles to cytoscape.js styles. I expect the reverse translation to be deployed by summer. Further, I also expect the cytoscape.js SBGN visualization to be integrated with the NDEx network display app in the timeframe.
Dexter Pratt Director, Software Development (Cytoscape, NDEx) Ideker Lab
marcoxa
Background
The BIMIB and the SYSBIO groups at the Universita` degli Studi di Milano-Bicocca have constructed a series of algorithms and tools to study the Metabolic Rewiring of Cancer cells. The main tool embodying such studies leverages the Matlab COBRA Toolbox to perform a Flux Balance analysis coupled with Linear Programming optimizations to highlight the different behaviors of Cancer cells w.r.t. normal cells. The "Metabolic Reaction Enrichment Analysis" (MAREA) tool computes a score for each relevant reaction, and uses it to provide a direct feedback to the user about how a reaction is "up-regulated" or "down-regulated", according to a given set of external constraints. At the time of this writing, the MAREA tool produces a static, highlighted version of the differentially enriched pathways by means of a set of Matlab and Python programs.
Goal
In order to improve MAREA usability and its visualization capabilities of the enriched metabolic pathways, we would like to propose to the development of a cytoscape.js based app, named MAREA/JS. This could be a self contained GSoC project for a motivated programmer. The main goal will be to read in the original metabolic network (which is usually a "reduced" one, w.r.t. the cell's full blown metabolic network) with all its annotations, either from a small data store, or, possibly, over a new MATLAB-cyREST bridge, and to display it in the main cytoscape.js window. Next the MAREA/JS UI will be used to explore the enriched pathways corresponding to the metabolic rewiring. The cytoscape.js display will also allow for the linking of web-based resources (e.g., gene annotations), in order to provide more information for the user.
Difficulty level: 2
Prior experience with cytoscape.js app development or extensive Javascript experience highly recommended. Python and MATLAB also useful as well as knowledge of SBGN as a possible exchange format.
Skills
Cytoscape.js, Javascript, Node.js, Matlab, COBRA Toolbox.
References
Graudenzi et al., Integration of transcriptomic data and metabolic networks in cancer samples reveals highly significant prognostic power, Journal of Biomedical Informatics, 87, 37-49, 2018.
Potential Mentors
Marco Antoniotti Alex Graudenzi
Contact
Marco Antoniotti (mailto:marco.antoniotti@unimib.it)