I have been developing network visualization implementations for Python, mostly because I needed them. I also believe in open source principles, that broadly useful and applicable code should be made freely available with their implementations made transparent. Based on these principles, I have released Python implementations of circos and hive plots, with a ton of help from Justin Zabilansky (MIT, now AeroFarms) and Jon Charest (MGH, who, by the way, is an aspiring data scientist).
At SciPy 2016, I had the opportunity to chat with our conference co-chair, Aric Hagberg, who also happens to be the lead maintainer of the NetworkX package. Specifically, I asked about contributing circos and hive plots into NetworkX. From our discussion, and from discussing with Jon, we have this idea to create an
nxviz package that provides a consistent API for creating network visualizations. Really excited about this! I also roped in a brilliant 1st-year UW undergrad Nelson Liu, who also has some network visualizations that he’d like to implement for directed acyclic graphs (DAGs). (No pressure, of course!)
Looking forward to developing this "rational" network visualization package!