Related Experiment Video
Updated: Jun 20, 2026
05:47
Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems
Published on: June 13, 2025
A Novel Approach to Zero-Shot Drug-Drug Interaction Prediction Enabled by EHR-Augmented Knowledge Graphs
Srijith Chinthalapudi1, Sandeep K Mallipattu2, Alisa Yurovsky2
1Edison Academy Magnet School, Edison, NJ.
Summary
This study introduces a new method to predict adverse drug-drug interactions (DDIs) by combining electronic health records (EHRs) with knowledge graphs (KGs). This approach enhances zero-shot DDI prediction for previously unseen drugs.
Area of Science:
- Pharmacovigilance and Biomedical Informatics
Background:
- Screening for adverse drug-drug interactions (DDIs) is a critical pharmacovigilance task.
- Electronic Health Record (EHR)-based statistical methods suffer from high false positive rates.
- Knowledge Graph (KG)-based methods lack zero-shot prediction for new drugs.
Purpose of the Study:
- To develop a novel approach for zero-shot DDI prediction.
- To augment incomplete biomedical KGs with real-world EHR data.
- To enable accurate prediction of DDIs for drugs not present in the initial KG.
Main Methods:
- Augmenting large-scale biomedical KGs with EHR-derived edges.
- Utilizing EHR associations as bridges to connect unseen drugs to KG knowledge.
- Designing a KG-embedding experiment isolating drugs during training for rigorous testing.
Main Results:
- The proposed method effectively enables zero-shot DDI prediction.
- EHR-derived associations bridge the gap for unseen drugs in KGs.
- Quantitative results demonstrate the specific efficacy of the approach for zero-shot DDI prediction.
Conclusions:
- Combining EHR data with KGs overcomes limitations of individual methods.
- The novel approach successfully achieves zero-shot DDI prediction capabilities.
- This method represents a significant advancement in pharmacovigilance and drug safety.
Related Concept Videos
01:29Pharmacogenomics: Identification of New Drug Targets
Advances in genomics have profoundly influenced drug discovery by increasing both the speed and accuracy of pharmaceutical development. Pharmacogenomics, which examines how genetic variation influences drug response, facilitates the identification of novel therapeutic targets and enables patient stratification for personalized treatment. These strategies contribute to improved drug efficacy, minimized adverse effects, and more efficient clinical trial design.Mapping genetic differences...
01:25Pharmacokinetics: Drug–Drug Interactions
Drug interactions occur when the pharmacological effect of one drug is altered by another substance, either enhancing or diminishing its activity. The drug whose activity is altered is known as the object drug, and the substance causing the alteration is called the agent drug or the precipitant. The net effects of these interactions are mostly undesirable, leading to decreased effectiveness or increased adverse effects. In rare cases, interactions can be beneficial, such as the enhanced...
01:30Drug toxicity: Drug–Drug Interaction
Drug–drug interactions can precipitate toxicity through multiple mechanisms. Absorption interactions alter how drugs enter the body, exemplified when ranitidine increases the absorption of basic drugs, while cholestyramine decreases the levels of propranolol. Protein binding interactions occur when drugs share the same binding sites on plasma proteins. Drugs like aspirin and warfarin, when bound in excess, can lead to increased free drug concentrations, enhancing the potential for...
01:26Drug Discovery: Overview
Drug discovery is a multifaceted process involving extensive screening, testing, and optimization of lead compounds to identify potential new drugs for therapeutic use. It combines several approaches, including screening large numbers of natural products, chemical modification of known active molecules, identification of new drug targets, and rational design based on biological mechanisms and drug-receptor structure. These approaches are carried out in both academic research laboratories and...
01:06Model Approaches for Pharmacokinetic Data: Distributed Parameter Models
Pharmacokinetic models are mathematical constructs that represent and predict the time course of drug concentrations in the body, providing meaningful pharmacokinetic parameters. These models are categorized into compartment, physiological, and distributed parameter models.
The distributed parameter models are specifically designed to account for variations and differences in some drug classes. This model is particularly useful for assessing regional concentrations of anticancer or...
The distributed parameter models are specifically designed to account for variations and differences in some drug classes. This model is particularly useful for assessing regional concentrations of anticancer or...
01:30Quantitative Aspects of Drug-Receptor Interaction
The receptor occupancy theory connects a drug's response to the number of occupied receptors. With higher drug concentrations, more receptors are occupied, leading to increased responses. The formation of drug-receptor complexes involves association and dissociation rates, which reach equilibrium when the forward and backward reactions are equal. The equilibrium association constant (Ka) and its inverse, the equilibrium dissociation constant (Kd), indicate drug affinity. Higher Ka and lower Kd...