Smart Drug Drug Interaction Detector by Mayo Clinic - UMKCNSF/UMKC--HACKATHON GitHub Wiki

Use case id: 8-MA-DrugInteraction

Title: Smart Drug-Drug Interaction Detector by Mayo Clinic

Attendees: Feichen Shen from Mayo Clinic

Presenter: Matthew Velazquez

Problem statement:

 •Drug-Drug interaction may decrease how well your medications work, may increase the side effects, or even increase the blood level and possible toxicity

 •Find drug-drug interactions have a significant meaning in clinical practice

 •Many drug-drug interactions are hidden in free-text literature

Objective:

Implement a Smart DDI system that is able to mine drug-drug interaction from PubMed literature.

Functions

1. Named Entity Recognition

    •Detect drug name from literature

    •ChemSpot: https://github.com/rockt/ChemSpot

Example:

2. Drug Name Mapping

   . Map drug name to DrugBank ID

   . DrugBank API: https://www.drugbankplus.com/

Example:

3. Visualize the drug network

 .Link different drugs mined from one or more pieces of literature by co-occurrence, you can use nodes to represent each drug and edges to represent their associations

 .d3.js can be used

Example

4. Drug-Drug Interaction

.Co-occurrence of drugs in one work of literature cannot guarantee the positive interaction between them. Try to figure out which two drugs have real interactions

 .Drug interaction API:

         https://rxnav.nlm.nih.gov/InteractionAPIs.html#

Example

Note 👍 Additional points for innovation

Questions:

Please create an issue on Github with usecase-id or email to: [email protected]