AI System Identifies Host Protein Targets for Broad-Spectrum Antivirals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for addressing viral diseases, such as COVID-19, face challenges in developing effective treatments due to the emergence of new strains and the potential for drug resistance, highlighting the need for broad-spectrum antiviral strategies that target host proteins rather than viral enzymes.
Innovation Solution
The system identifies therapeutic targets by calculating a differential interaction score (DIS) based on protein-protein interactions, correlating it with the likelihood of dysfunctional interactions being causal agents of disorders, and selecting appropriate treatments, which includes administering specific inhibitors like PGES-2 or sigma receptor inhibitors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional antivirals target viral enzymes, then antiviral activity is achieved, but drug resistance develops due to viral mutation
Solution Approach 1:
Instead of targeting viral enzymes (traditional approach), the patent inverts the strategy by targeting host proteins that are essential for viral replication. This approach targets the host cell's machinery that the virus exploits, making it difficult for the virus to develop resistance without losing its ability to replicate.
Solution Approach 2:
The patent identifies host proteins that serve multiple functions - they are required for both normal cellular processes and viral replication. By targeting these universal host factors, a single therapeutic can potentially address multiple coronavirus strains and prevent resistance development, as the host protein cannot be easily mutated without compromising cellular function.
2Adaptability or versatility
If host proteins are targeted for broad-spectrum activity, then resistance barrier is increased, but identification of specific therapeutic targets becomes more complex
Solution Approach 1:
The patent replaces traditional experimental and observational methods with an artificial intelligence-based computational system. The AI analyzes protein interaction networks, predicts essential host factors for viral replication, and prioritizes therapeutic targets, thereby simplifying the complex process of identifying broad-spectrum antiviral targets.
Solution Approach 2:
The patent introduces an AI-based computational framework as an intermediary between the complex biological system (host-virus interactions) and the therapeutic development process. This intermediary system processes vast amounts of proteomics and interactomics data to identify and prioritize host proteins that are essential for viral replication, making the target identification process more systematic and less complex.
3Measurement precision
If protein-protein interaction networks are analyzed to identify causal agents, then treatment precision is improved, but computational and analytical complexity increases
Solution Approach 1:
The patent replaces manual biochemical analysis and traditional experimental methods with an AI-based computational system that automatically analyzes protein-protein interaction networks. The AI processes complex interactomics and proteomics data to identify causal dysfunctional interactions, thereby improving treatment precision while managing computational complexity through automated algorithms.
Data Source
AI summary
The disclosure relates to a system comprising software that predicts responsiveness of subjects to certain disease modifying drugs. Embodiments of the disclosure include methods comprising calculating a differential interaction score (DIS), correlating the DIS with the likelihood that a dysfunctional protein-protein interaction is the causal agent of a disease or disorder, and identifying a subject responsive to a treatment based upon the causal agent.