kana is a web application for interactive analysis of single-cell genomics data. Its unique selling point is that all computation is done on the client machine, rather than being transferred to a backend server; this simplifies deployment, avoids latency and data ownership issues, and provides greater scalability across users. Check out our manuscript for more details.
This organization manages the code for the kana application as well as the underlying packages and resources. Most code is written in Javascript, with a sprinkling of C++ for WebAssembly and R for some preprocessing.
bakana ("backend kana") manages the analysis pipeline and results for the web application. It can also be used to execute the same analysis on a backend server via Node.js.
scran.js implements various methods for single-cell data analysis, by compiling various C/C++ libraries such as libscran to WebAssembly.
kanapi provides a more traditional backend for kana, allowing maintainers to run bakana on a server and transfer results to the client via websockets.
This organization is managed by Aaron "A-bomb" Lun and Jayaram "Rowling" Kancherla.