Graph embedding-based link prediction for literature-based discovery in Alzheimer’s Disease

Please cite this article as: Y. Pu, D. Beck and K. Verspoor, Graph embedding-based link prediction for literature-based discovery in Alzheimer’s Disease, Journal of Biomedical Informatics (2023), doi: https://doi.org/10.1016/j.jbi.2023.104464.

Prerequisites

• Python 3.8

Directory Structure

• ./00_data directory contains ontologies and metadata
• ./01_corpus directory contains the list of pmids used in the study
• ./02_collect_corpus directory contains scripts for collecting corpus
• ./03_create_annotator directory contains scripts for creating AD-specific annotators
• ./04_process_annotation directory contains scripts for processing annotations
• ./05_generate_graph directory contains scripts for generating the AD knowledge graph
• ./06_infer_knowledge directory contains scripts for predicting putative links with graph embedding models
• ./07_analyze_graph directory contains scripts for analyzing graph statistics
• ./08_analyze_predictions directory contains scripts for analyzing outputs from link prediction models

Contact

Karin Verspoor (karin.verspoor (at) rmit.edu.au) or Yiyuan Pu (yiyuanp1 (at) student.unimelb.edu.au)