This repository is a part of the master's thesis "Graph-based representation, integration, and analysis of neuroscience data — The case of the murine basal ganglia".
In this thesis, we investigate aspects of graph-based data representation in the neuroscience domain.
The data set basis is a relational database published by Bjerke et al. (2019) of quantitative neuroanatomical data about the healthy rat and mouse basal ganglia (DOI:10.25493/DYXZ-76U). This repository contains the relational database as CSV-files, located at /Data/csvs/basal_ganglia. The illustration below presents the high-level solution design of the master's thesis. The project's primary notebook converts and inserts the data from these CSV-files into a Neo4j graph database instance. In addition, this project contains notebooks for extending data from external sources and for analyzing the data in the graph model (requires Neo4j's Graph Data Science Library).
Run the following commands to install the required packages:
$ pip3 install -r requirements.txt
$ jupyter notebookAll the notebooks use dotenv to store connection credentials. The driver connection requires the Neo4j username, password, and bolt URL to run the notebooks.
Following the solution design, we have separated the different aspects of this project. At the root lies the notebook that creates the graph database from the relational
(1. Create the Rodent Basal Ganglia Graph.ipynb).
The folder 1. Extending the dataset with data from other sources presents the efforts in collecting data from external sources. Here, we present how we collect data from NeuroMorpho.Org and InterLex through the SciCrunch API. For full citation of the downloaded morphologies, see the neuromorpho data folder README.
The folder 2. Extending the data with extracted information - GraphAnalysis presents the efforts for running graph algorithms on the data. The notebooks 1. General information search.ipynb and 1. General information search.ipynb presents the exploratory data analysis, using the Neo4j Graph Data Science Library and NetworkX, respectively. The notebook 3. UseCases.ipynb presents the confirmatory data analysis, where we apply analytics to answer specific questions about the data.
I. E. Bjerke et al. "Database of quantitative cellular and subcellular morphological properties from rat and mouse basal ganglia [Data set]". In: *Human Brain Project Neuroinformatics PlatformØ (2019). DOI: 10.25493/DYXZ-76U.
Giorgio A Ascoli, Duncan E Donohue, and Maryam Halavi. "NeuroMorpho. Org: a central resource for neuronal morphologies". In: Journal of Neuroscience 27.35 (2007), pp. 9247–9251.
FDI Lab - SciCrunch Infrastructure | InterLex | Dashboard. https://scicrunch.org/scicrunch/interlex/dashboard. Last accessed: 2020-08-30