/project-symptoms-disease-network

Analysis of the Symptoms-Disease Network database using communities.

Primary LanguageJupyter NotebookMIT LicenseMIT

made-with-python MIT Licence Ask Me Anything ! Open Source Love Tweet

Project Symptoms-Disease Network

Overview

Analysis of the Symptoms-Disease Network database using communities.

Usage

The complete code of this project is in this repository, in the file: symptoms-disease-network.ipynb.

If the notebook does not load or takes a long time to load, please refresh the page.

Approach

The steps developed in this project were as follows:

figure

Dataset

The dataset was extracted from the paper: "Human symptoms–disease network" Link.

More exactly, here we employ the "Supplementary Data 3": Term co-occurrences between symptoms and diseases measured by TF-IDF weighted values. This table includes 147,978 records of symptom and disease relationships.

figure

All Networks: bipartite graph and their projections

figure

Communities of Diseases Network using Louvain

figure

Bipartite graphs of each community by symptoms and diseases

figure

figure

Table of the characterization of each community by symptoms

figure

Predict the disease category based on symptoms

figure

References

  • Zhou, X., Menche, J., Barabási, AL. et al. Human symptoms–disease network. Nat Commun 5, 4212 (2014). https://doi.org/10.1038/ncomms5212
  • Blondel, V. D., Guillaume, J.-L., Lambiotte, R., & Lefebvre, E. (2008). Fast unfolding of communities in large networks. Journal of Statistical Mechanics: Theory and Experiment, 2008(10), P10008. doi:10.1088/1742-5468/2008/10/p10008