Today is my third day in Dr. Bassett's lab and I began to learn how to use BrainNet and Gephi. I will be spending the first week getting used to these two programs.
Both BrainNet and Gephi are programs that help people to visualize networks. BrainNet has brain models (transparent gray brain images in the figure below), and when you import data, the program automatically applies the data to the brain models. BrainNet can create realistic three dimensional representations of neural networks. Dr. Bassett gave me a bunch of data that I can play around with, and the image below is one of the outputs that I got when I applied those data to BrainNet.
Unlike BrainNet which is only used for neural networks, Gephi can be used to visualize any types of networks. If you have time, you should try using Gephi to get a visual representation of your facebook friend network (here is the link if you want to see a video tutorial for it: https://youtu.be/kbLFMObmLNQ).
Gephi seemed overwhelming at first, but now I am having fun playing with it. These three images above are all produced with same data, but they have different layouts and settings. Each dot in the images represents a node, and each line represents a neural connection. For the first image (left), I grouped the nodes based on their connectivity (relatively cliquey nodes become a group) and I color coded the groups. In order to differentiate more popular nodes (with more connections) from less popular nodes, I made the nodes in the second image (middle) have different sizes (the more popular ones are larger). For the third image (right), I simplified the image by getting rid of some unpopular nodes that have less than ten connections. I still have a lot left to learn but i think I will enjoy the process.
No comments:
Post a Comment