Probabilistic Pocket Ensemble for Drug Target Prediction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for predicting drug targets are limited in capturing structural flexibility and conformational variations, leading to low sensitivity and specificity in identifying off-target binding sites, which contributes to drug toxicity and inefficacy.

Innovation Solution

A novel system-based approach integrating sequence order-independent structure alignment, divisive hierarchical clustering, and probabilistic sequence similarity techniques to construct a probabilistic pocket ensemble (PPE) that captures promiscuous structural features of drug binding sites, combined with an approximation of drug delivery profiles for large-scale prediction of novel drug-protein interactions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional structure-based methods are used for target prediction, then the method is simple to implement, but the sensitivity and specificity in identifying off-target binding sites remain low

Engineering Contradiction:
Improveprediction accuracyVSAvoidmethod complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent combines structure-based methods (binding site similarity, molecular docking) with expression-based methods (molecular activity perturbation signatures) and ligand-based methods (chemical and structural properties) into an integrated framework. This merging of multiple prediction approaches enables the system to capture both on-target and off-target binding sites with higher sensitivity and specificity while maintaining practical implementability through a unified computational pipeline.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The invention creates a composite prediction model that integrates multiple types of biological and chemical data (protein structures, gene expression profiles, ligand properties) into a unified target prediction system. This composite approach leverages the strengths of different methodological categories to achieve superior prediction accuracy compared to any single method alone.

Inventive Principle:
Principle #40Composite materials

2Adaptability or versatility

If current structural information methods are used, then the computational approach is straightforward, but the structural flexibility and conformational variations of drugs are not captured

Engineering Contradiction:
Improvestructural flexibility captureVSAvoidcomputational framework complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent employs molecular dynamics simulations and conformational sampling techniques to capture the dynamic behavior of drugs and their targets. By modeling multiple conformational states and flexible binding modes, the system can identify off-target binding sites that involve conformational changes, thereby improving adaptability to structural flexibility while maintaining a manageable computational framework through efficient sampling methods.

Inventive Principle:
Principle #15Dynamics

3Reliability

If comprehensive target prediction is performed to understand off-target effects, then drug toxicity and inefficacy can be reduced, but the computational resources and time required increase

Engineering Contradiction:
Improvedrug efficacy and safetyVSAvoidprediction time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent performs preliminary filtering and prioritization of potential off-target binding sites using rapid computational screens before conducting more detailed analysis. By pre-identifying high-probability targets based on structural and expression data, the system reduces the number of candidates requiring extensive validation, thereby improving drug reliability while minimizing time loss in the overall prediction process.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The invention implements a tiered prediction approach where a comprehensive initial screen identifies all potential targets, followed by focused analysis on the most promising candidates. This partial action strategy ensures that sufficient coverage is achieved to identify critical off-target effects while avoiding unnecessary computational expenditure on low-priority targets, thus balancing reliability improvement with time efficiency.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3839061B1Methods and systems for identifying ligand-protein binding sites
Publication Date: 2022.12.07 KING ABDULLAH UNIV OF SCI & TECH
  • EP3839061B1 patent drawingFigure 1a~2b
  • EP3839061B1 patent drawingFigure 3a~3d
  • EP3839061B1 patent drawingFigure 4a~4d

AI summary

The invention provides a novel integrated structure and system-based approach for drug target prediction that enables the large-scale discovery of new targets for existing drugs. Novel computer-readable storage media and computer systems are also provided. Methods and systems of the invention use novel sequence order-independent structure alignment, hierarchical clustering, and probabilistic sequence similarity techniques to construct a probabilistic pocket ensemble (PPE) that captures even promiscuous structural features of different binding sites for a drug on known targets. The drug's PPE is combined with an approximation of the drug delivery profile to facilitate large-scale prediction of novel drug-protein interactions with several applications to biological research and drug development.