https://www.s-cubed-global.com/news/covidgraph-nerds-response-to-the-pandemic
https://github.com/covidgraph/documentation/issues/8
This python script loads data from ClinicalTrials.gov API into the neo4j based covidgraph. The script gets data from the StudyFields, which are described on this API homepage.
Maintainer: Kirsten
Version: 0.2.1
Neo4j version: < 3.5.17
APOC version: < 3.5.0.11
Docker image location: covidgraph/data-clinical_trials_gov
docker run -it --rm --name data-cord19 --network host -e NEO4J='{"host":"localhost"}' covidgraph/data-clinical_trials_gov
docker build -t data-clinical_trials_gov .
docker run -it --rm --name data-cord19 --network host -e NEO4J='{"host":"localhost"}' data-clinical_trials_gov
The most important Env variables are:
NEO4J: defaults to {"host":"localhost"}. The connections details for the database. For details see https://github.com/covidgraph/motherlode/blob/master/README.md#the-neo4j-connection-string
To run the code without docker you need to have python installed.
Setup
Install the python requirments with
pip install --no-cache-dir -r requirement.txt
Run the script with
python3 ./dataloader/main.py
Due to a limit of 1000 studies to be returned from a query[https://clinicaltrials.gov/api/gui/demo/simple_study_fields], the queries has been split into 3 parts (syntax for the query): Studies contatining the word COVID for
- Obervational studies (COVID AND AREA[StudyType]Observational)
- Interventional studies (COVID AND AREA[StudyType]Interventional)
- NOT (Observations AND Interventional) studies - e..g expanded access(COVID AND NOT AREA[StudyType]Interventional AND NOT AREA[StudyType]Observational)
Decription of the fields can be found here: https://clinicaltrials.gov/api/gui/ref/crosswalks.
At this point no results information can be found for COVID studies. This will be added once results are avilable.
