Neoantigen Prediction via AI Molecular Dynamics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current anticancer immune vaccines based on neoantigens face challenges of high cost and relatively low efficiency due to the diverse types and number of MHC proteins and the need for efficient identification of immunogenic neoantigens in each patient.
Innovation Solution
A molecular dynamics-based system (NeoScan) using AI-based big data analysis to predict neoantigen binding affinities for MHC, identifying patient-specific neoantigens through genomic mutations, and inducing immune responses using TCR-T cells, CAR-T cells, and TILs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If bioinformatic neoantigen prediction technologies based on genomic information are used, then the opportunity for identifying neoantigens increases, but the cost remains high and efficiency is relatively low
Solution Approach 1:
The patent segments the neoantigen identification process into multiple stages: initial genomic mutation screening, followed by AI-based molecular dynamics simulation for binding affinity prediction, and finally experimental validation. This segmentation allows each stage to focus on specific tasks, improving overall efficiency while maintaining high predictive accuracy through the integration of multiple methods.
Solution Approach 2:
The patent introduces AI-based molecular dynamics simulation as an intermediary step between genomic information analysis and final neoantigen selection. This intermediary uses deep learning models trained on molecular dynamics data to predict binding affinities, thereby enhancing the reliability of neoantigen identification without requiring extensive experimental screening of all candidates.
2Measurement precision
If comprehensive MHC protein typing is performed for each patient, then personalized neoantigen identification accuracy improves, but the complexity and cost of the process increases
Solution Approach 1:
The patent develops a universal AI-based molecular dynamics prediction system that can handle multiple MHC alleles and neoantigen candidates simultaneously. The deep learning model is trained on diverse MHC-peptide binding data, enabling it to predict binding affinities across different MHC types without requiring separate analysis pipelines for each allele, thus reducing complexity while maintaining precision.
Solution Approach 2:
The patent changes the approach from direct MHC typing to predicting binding affinity parameters through AI models. Instead of performing comprehensive MHC typing and then matching with neoantigens, the system directly predicts binding affinity parameters using molecular dynamics simulations and deep learning, simplifying the process while improving accuracy through physics-based calculations.
3Productivity
If AI-based molecular dynamics big data is used for neoantigen prediction, then predictive ability and efficiency are enhanced, but the computational resources and data processing requirements increase
Solution Approach 1:
The patent performs preliminary actions by pre-training deep learning models on extensive molecular dynamics simulation data before actual neoantigen prediction. The AI models are pre-trained on databases of MHC-peptide binding affinities generated through molecular dynamics simulations, allowing them to make rapid predictions without requiring real-time computational resources during the actual screening process.
Solution Approach 2:
The patent uses copying by creating simplified representations of complex molecular interactions through AI models. Instead of performing full molecular dynamics simulations for each neoantigen candidate, the system uses pre-trained deep learning models that copy the essential binding affinity predictions from the training data, dramatically reducing computational resource requirements while maintaining prediction accuracy.
Data Source
AI summary
Disclosed are a system and a method of predicting neoantigens and immune response induction. The system and the method may verify induction of immunity against neoantigens having high binding affinity by identifying neoantigen candidates through genomic mutations and then predicting the binding affinities of the neoantigen candidates for MHC through molecular dynamics. The method provides neoantigen immunotherapy information for identifying a neoantigen using artificial intelligence (AI)-based molecular dynamics big data, and includes steps of: (A) identifying neoantigen candidates through a genomic mutation; (B) filtering the specificities of the neoantigen candidates for tissue and disease; (C) predicting the in silico binding of the neoantigens to MHC; and (D) calculating and ranking TCR activity.