Drug-Target Interaction Prediction Using Multi-Attribute Association Matrices

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for predicting drug-target interactions using computers are inefficient and inaccurate due to the inability to accurately extract feature information of drugs and target proteins, leading to low accuracy in predicting interactions.

Innovation Solution

A method involving determining a first drug association matrix and a first target association matrix based on drug and target attribute information, using graph convolution models to characterize feature information, followed by a fusion model to predict interaction probabilities.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If experiments are used to determine interaction between drug and target protein, then reliability of interaction determination is improved, but productivity is worsened due to inefficiency

Engineering Contradiction:

Improvereliability of interaction determinationVSAvoidproductivity of interaction determination

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent replaces the mechanical experiment system with a computer-based prediction system that uses graph convolution models and association matrices to predict drug-target interactions, thereby eliminating the need for physical experiments while maintaining reliability

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent creates a virtual model (association matrix) that copies and represents the complex relationships between drugs and targets, allowing prediction of interactions without physical experimentation

Inventive Principle:

Principle #26Copying

2Productivity

If computer prediction methods are used to predict drug-target interaction, then productivity is improved, but measurement precision is worsened due to inability to accurately extract feature information

Engineering Contradiction:

Improveproductivity of interaction predictionVSAvoidaccuracy of feature information extraction

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The patent segments feature extraction into multiple independent association matrices (drug association matrix, target association matrix, interaction association matrix) that can be processed separately and then integrated, improving both precision and efficiency

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent transforms feature information into multi-dimensional association matrices that capture relationships from multiple angles (drug-drug, target-target, drug-target interactions), enabling more accurate extraction and prediction

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

3Device complexity

If simple computer prediction models are used, then device complexity is reduced, but measurement precision is worsened due to low accuracy in predicting interaction

Engineering Contradiction:

Improvecomplexity of prediction modelVSAvoidaccuracy of interaction prediction

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent combines multiple types of association matrices (drug association matrix, target association matrix, interaction association matrix) into a composite prediction model that leverages the strengths of each component to achieve high accuracy

Inventive Principle:

Principle #40Composite materials

Data Source

PatentUS20260024612A1Method and device for predicting drug-target interaction, and storage medium

Publication Date: 2026.01.22 BOE TECHNOLOGY GROUP CO LTD

AI summary

A method for predicting drug-target interaction includes: determining a first drug association matrix according to drug attribute information, the drug attribute information including at least one of a drug structure similarity, a pharmacophore similarity, a side effect similarity, and a GO pathway-based similarity of drugs, and the first drug association matrix being used to characterize feature information of each drug on at least one drug attribute; determining a first target association matrix according to target attribute information, the target attribute information including at least one of a target structure similarity and a target interaction relationship of targets, and the first target association matrix being used to characterize feature information of each target on at least one target attribute; and predicting a probability of interaction between a drug and a target according to the first drug association matrix and the first target association matrix.