/clinical_trials_3000

Primary LanguageJupyter NotebookMIT LicenseMIT

Introduction

This repository contains code and data for my article "Getting Insights from 3,000+ Clinical Trials in a Knowledge Graph".

  1. The Python scripts are for the chatbot.

Prerequisite

Neo4j Desktop or AuraDB

GCP account

UMLS free tier

Enbale Vertex AI API in the GCP project, increase its quota (default 5)

501 VECTOR INDEX is not available in granular instances in preview 12: VECTOR INDEX is not available in granular instances in preview

Run

  1. Set up the databases as described in the article.

  2. Install the necessary Python libraries (install.txt) and gcloud CLI

  3. Fill the necessary API key information in config.yaml

  4. Run

clinical_trials_to_kg.ipynb

and

get_meta_data_experiment.ipynb

to convert the clinical data to TSV

  1. Import the data into Neo4j

  2. Run

to_bigquery.ipynb

to export data to GCP and to create dashboard

Authors

  • Sixing Huang - Concept and Coding

License

This project is licensed under the MIT License - see the LICENSE file for details