Neoepitope Payload Toxicity Prediction for Therapeutic Viruses
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining recombinant viral payload toxicity in therapeutic viruses are inadequate, particularly for artificial sequence constructs like neoepitopes, as toxicity is often only evident after large-scale production, leading to reduced expression and potential harm to patient cells.
Innovation Solution
A method involving the expression of recombinant nucleic acid sequences in host cells, sequencing, and correlating sequence portions with toxicity measures using machine learning classifiers, such as linear and autoencoders, to predict and mitigate payload toxicity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If recombinant nucleic acid sequences are expressed in host cells for large-scale virus production, then viral titers are achieved, but payload toxicity becomes evident causing reduced expression and potential harm to patient cells
Solution Approach 1:
The patent applies preliminary action by performing toxicity assessment through sequencing and machine learning analysis before large-scale virus production begins. The method sequences recombinant nucleic acid sequences early in the development process and uses computational models to predict toxicity, allowing researchers to identify and mitigate potential toxicity issues before they affect patient cells during therapeutic application.
2Measurement precision
If toxicity assessment is performed using predictive algorithms on naturally occurring polypeptides, then toxicity can be identified, but the method is not applicable to artificial sequence constructs like neoepitopes
Solution Approach 1:
The patent applies parameter changes by modifying the input parameters and training data of machine learning models to accommodate artificial sequence constructs. Instead of using only naturally occurring polypeptide data, the methodology trains predictive algorithms on diverse sequence types including neoepitopes and synthetic constructs, adjusting the computational parameters to recognize patterns in artificial sequences while maintaining toxicity prediction accuracy.
3Productivity
If expression of recombinant payload is suppressed in production cells to avoid toxicity, then high viral titers are achieved, but neoepitope expression is reduced in patient cells
Solution Approach 1:
The patent applies preliminary action by using sequencing and machine learning-based toxicity prediction to identify potentially toxic payloads before production. This allows researchers to select or engineer alternative payload sequences that maintain immunogenicity while reducing toxicity risk, thereby avoiding the need to suppress expression and ensuring both high viral titers and adequate neoepitope expression in therapeutic applications.
Data Source
AI summary
Systems and methods are presented that allow for determination and prediction of payload toxicity in therapeutic viruses. Disclosed herein are methods of determining payload toxicity of an expressed polypeptide in a cell, comprising: generating or procuring a plurality of expression vectors, each containing a different recombinant nucleic acid sequence that encodes a corresponding recombinant polypeptide; expressing the recombinant nucleic acid sequence in a plurality of host cells while culturing the host cells; sequencing the plurality of expression vectors after culturing the host cells; and correlating at least portions of the recombinant nucleic acid sequence with a toxicity measure.