KIDS constructs an inconsistency-free knowledge graph that supports multiple triple types and performs knowledge graph completion. We apply the KIDS framework to the area of Eschericia coli antibiotic resistance. This work proposes an integrated approach towards automated knowledge representation and discovery, and demonstrates how evidence-driven decisions can automate knowledge discovery with high confidence and accelerated pace.
Figure 1. Overview of the KIDS framework.
./hypothesis_generator: Code for generating the hypothesis based on the knowledge graph created. Please refer to its own README file for more information../images: Contains README file related images../kg_constructor: Code for creating the knowledge graph. Note that this directory also contains the source code for performing inconsistency resolution. Please refer to its own README file for more information../manuscript_preparation: Code for analyzing the results as reported in the manuscript.
This code has been tested with Python 3.6 under both Ubuntu 18.04 LTS and Ubuntu 20.04 LTS.
mkdir KIDS
git clone https://github.com/IBPA/KIDS.git ./KIDS
Create and activate virtual environment.
cd ./KIDS
python3 -m venv env
source env/bin/activate
Install all required python packages once the virtual environment has been activated.
pip3 install -r requirements.txt
You will also need Java 7 or higher. If you are running Ubuntu 18.04, follow the steps below to install Java OpenJDK 11.
sudo apt update
sudo apt install openjdk-11-jdk
(Optional) You can deactivate the virtual environment once finished.
cd ./KIDS
deactivate
- Construct the KG by following the README file.
- Generate the hypothesis by following the README file.
For any questions, please contact us at tagkopouloslab@ucdavis.edu.
We will update this section once citation information is available.
This project is licensed under the Apache-2.0 License. Please see the LICENSE file for details.
- Special thanks to the members of the Tagkopoulos lab and the reviewers for their suggestions.
- Nick Joodi and Minseung Kim for their help in the initial discussions, and Ameen Eetemadi for his comments on creating the figures.
- This work was supported by the USDA-NIFA AI Institute for Next Generation Food Systems (AIFS), USDA-NIFA award number 2020-67021-32855 and the NIEHS grant P42ES004699 to Ilias Tagkopoulos.