Chapter 6 called Graph Theory in my book 'A handbook of mathematical models with python' talks about graphs, graph-structured data, and how they serve as inputs to graph neural networks (GNNs).
Buy book from Amazon: https://a.co/d/7Yz0usb
Studying a protein with an elastic network model that includes coarse-grained Gaussian network model (GNM) and atomic anisotropic network model (ANM):
https://www.pnas.org/doi/full/10.1073/pnas.0902159106
https://www3.mpibpc.mpg.de/groups/de_groot/pdf/Hayward_deGroot_nm_ed.pdf
-> Elastic network model (coarse-grained model) to study protein dynamics:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6320916/
-> My work on anisotropic network model:
Wikipedia reference: https://en.wikipedia.org/wiki/Anisotropic_Network_Model
Studying protein dynamics with python: http://prody.csb.pitt.edu/tutorials/enm_analysis/
OTHER USE CASES
-> Find most optimal (flight) routes in terms of distance & airtime using Dijkstra algorithm from (weighted) graphs
More on Dijkstra algorithm for a graph geodesic:
https://mathworld.wolfram.com/DijkstrasAlgorithm.html
-> Create knowledge graphs (directed) from unstructured data (document, webpage, etc.):
https://colab.research.google.com/drive/1EF_ra7u6qHqG5p5vmYDYC9X5Y06hsub7?usp=sharing
-> Do social network analysis with graphs from your data of connections/contacts on social site
-> There are operations research problems (routing etc.) that can be solved utilizing Network Science.