Closed giovanniscardoni closed 5 years ago
Hello,
I don't have any prior experience with Cytoscape app. Can I still apply for this project?
I'm confortable with the AI algorithms referenced in the text and with some others.
Thanks!
Hi, Welcome to Cytoscape!
To get familiar with Cytoscape and its apps, Please download Cytoscape and tickle the software a little bit. Know about its features. I recommend you to follow this post[1]. Of course, it is a bit late now to work on bugs, but you can try out the software and know what the apps do. It is pretty simple and understanding if you know a little bit of Java. The Java documentation of Cytoscape is useful too[2].
Get Started with the App developer page [3]. Start from the 'getting started' section. Further you will need Cytoscape cookbook[4] to make much use of Cytoscape API.
Thanks and happy coding! Let us know your findings! Faizaan
[1] - https://www.quora.com/How-do-I-prepare-for-the-Google-Summer-of-Code-GSoC/answer/Srikanth-Bezawada [2] - http://chianti.ucsd.edu/cytoscape-3.3.0/API/ [3] - http://wiki.cytoscape.org/Cytoscape_3/AppDeveloper [4] - http://wiki.cytoscape.org/Cytoscape_3/AppDeveloper/Cytoscape_3_App_Cookbook
It would be amazing to join to this project... I know it's quite late to apply but could I do it?
Closing for GSoC 2019
Background
Complex networks are models which permit to represent interactions between very different objects. In biology, graphs are used in order to model interactions between molecular actors that are, for instance, genes, proteins and metabolites. Novel methodologies are making models more and more detailed and, due to this increasing complexity, it is becoming more and more difficult to extract useful informations, recurring pattern or similarities along network which belong to a specific class. There are a number of algorithms which are commonly used for tasks like automatic learning and classification, that allow searching for common characteristics which are shared along similar samples, e.g. a set of networks belonging to mammalian organisms versus a set of networks belonging to a set of bacteria.
Goal
CentiScaPe currently allows the creation of weighted networks that are used in order to model experimental dataset by using networks. Creating a set of networks which model a specific experimental condition and compare it with another set of networks which model a different experimental condition is now possible but the comparison of a number of network is not feasible without automatic models. The idea here is to extend the current version of CentiScaPe by adding some machine learning algorithms like, for instance, Linear Discriminant Analysis, K-Means, and K-Nearest-Neighbours. Ideally some graphical output will be useful in order to present the results.
Difficulty level: 3
Prior experience with Cytoscape app development, CenitScaPe and machine learning is required.
Technology and Skills
Cytoscape, CentiScaPe, Java, Pattern Recognition Algorithms
Potential Mentors
Giovanni Scardoni (CBMC, University of Verona)
Contact
cytoscape-discuss@googlegroups.com, giovanni.scardoni@gmail.com