Drug Repurposing via Pathway Frequency Weighting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current in-silico approaches for drug repurposing lack an informatics view of data related to drugs, pathways, and indications, failing to provide an effective mechanism for predicting the suitability of repurposed drugs for new therapeutic indications.
Innovation Solution
A method and system that calculates pathway weights based on the frequency of occurrence in positive and negative drug sets, generating a drug score for repurposed drugs by associating index values with drug actions on pathways, and using a clustering approach for scoring and prediction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If current in-silico approaches are used for drug repurposing, then the process is simpler and faster, but the prediction accuracy of drug fitment for new indications is insufficient
Solution Approach 1:
The patent segments the drug repurposing evaluation into multiple independent pathway analyses. Each pathway is scored separately based on drug-pathway associations, and these scores are aggregated to generate an overall drug fitment score. This segmentation allows for more comprehensive and accurate predictions while maintaining computational efficiency through modular processing.
Solution Approach 2:
The patent introduces a new dimension of analysis by incorporating pathway frequency information and pathway weights into the drug scoring mechanism. Instead of relying on traditional single-dimension drug-indication matching, the system evaluates drugs across multiple pathway dimensions, each weighted by its frequency and relevance, thereby enhancing prediction accuracy through multi-dimensional assessment.
2Measurement precision
If pathway frequency based modeling is implemented, then the drug fitment prediction becomes more accurate, but the computational complexity and data processing requirements increase
Solution Approach 1:
The patent performs preliminary actions by pre-calculating pathway frequencies and pathway weights from the training data before actual drug repurposing predictions are made. The pathway frequency based model is pre-trained and stored, allowing rapid inference during the actual drug scoring process without requiring intensive real-time computation, thus reducing the computational power needed during deployment.
Solution Approach 2:
The patent creates a simplified copy of the complex biological system through the pathway frequency based model. Instead of simulating all molecular interactions in real-time, the system uses pre-computed pathway scores and weights that replicate the essential predictive patterns, enabling accurate drug fitment prediction with significantly reduced computational requirements during the actual screening process.
3Reliability
If a clustering approach with alternate scoring mechanism is used, then the prediction mechanism becomes more robust, but the system complexity increases
Solution Approach 1:
The patent merges multiple scoring mechanisms into a unified pathway frequency based scoring system. By combining pathway frequency analysis, pathway weight calculation, and drug-pathway association scoring into a single integrated model, the system achieves robust and reliable predictions while avoiding the complexity of managing multiple separate scoring systems. The clustering approach naturally emerges from the data rather than being imposed as a separate complex mechanism.
Data Source
AI summary
The present invention provides a method and system for drug repurposing. In accordance with a disclosed embodiment, the method may include selecting a set of positive drugs for an indication and listing a set of pathways for the set of positive drugs and a set of negative drugs. Further, the method shall include associating a pathway for each drug with an index value, based on an action of the each drug on the pathway. A pathway weight shall be calculated from the index value of the each pathway based on a frequency of occurrence of the each pathway in the set of positive drugs and in the set of negative drugs. A drug score for a drug to be repurposed shall be identified from the negative set of drugs, based on the pathway weight. The drug with a relatively high drug score can be repurposed for the indication.