-
Notifications
You must be signed in to change notification settings - Fork 52
Description
Background
Cytoscape is an invaluable open-source tool in the field of biomedical research, offering robust data visualization capabilities, particularly for molecular networks. Its ecosystem, enriched by over 370 specialized apps available in the App Store, facilitates diverse functionalities ranging from database access to novel analysis methods. However, the recent development of Cytoscape Web, an online variant, has introduced complexities in maintaining the platform's integrity and quality across versions. As the ecosystem expands, the challenge of ensuring consistent and reliable product releases grows. Thus, the development of sophisticated testing tools becomes imperative to uphold the high standards of Cytoscape's offerings.
Goal
The primary goal of this project is to create an automated testing framework tailored for Cytoscape, with a specific focus on scientific figure preparation—a critical use case for the software. This framework will utilize a collection of Cytoscape session files and CX documents, integral in generating figures for various scientific publications. It aims to verify whether the newest versions of both Cytoscape and Cytoscape Web can accurately reproduce these figures. A key aspect of this project involves ensuring that figures produced by Cytoscape Web are consistent with those generated by the desktop version. The framework will facilitate communication with Cytoscape Desktop through the CyRest API and automate tests on Cytoscape Web using renowned frameworks like Playwright . A significant part of the testing process will involve comparing newly generated images against established 'gold standard' figures, employing image analysis tools (OpenCV, ImageMagick) and advanced language models (such as Gemini and GPT-4) to produce comprehensive test reports.
Difficulty Level: Medium
Size and Length of Project
- 175 hours
- 12 weeks
Skills
Javascript, TypeScript, Python, Cytoscape
Public Repository
Potential Mentors
- Jing Chen
- Kei Ono
- Dylan Fong
- Chris Churas