cfDNA Fragment Methylation and Size Profiling for Disease Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for predicting disease states using cell-free DNA (cfDNA) fragments are limited by analyzing only a single feature, leading to unreliable and computationally intensive processes, especially when the signal is weak, and fail to accurately distinguish circulating tumor DNA (ctDNA) from other DNA molecules.
Innovation Solution
Integrating multiple features such as methylation status and size of cfDNA fragments, including stratifying by size and comparing methylated loci within specific size ranges, to determine a disease score for improved predictive accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple features of cfDNA fragments are analyzed together, then predictive accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent segments the analysis by dividing cfDNA fragments into different size ranges (e.g., 100-150 bp, 150-200 bp, 200-250 bp) and analyzing methylation patterns separately for each size category. This segmentation allows the system to process multiple features (size and methylation) in a structured, computationally manageable way while maintaining high predictive accuracy for distinguishing ctDNA from cfDNA
Solution Approach 2:
The patent changes the parameter of analysis by focusing on specific size ranges and methylation levels rather than analyzing all cfDNA fragments uniformly. By adjusting these parameters (size thresholds, methylation cutoffs), the system achieves accurate disease state prediction with optimized computational resources
2Measurement precision
If computationally intense processes are used to analyze multiple cfDNA features, then predictive accuracy is improved, but ease of operation deteriorates
Solution Approach 1:
The patent replaces complex computational mechanics with a simplified interpretation framework. Instead of using black-box machine learning models that are difficult to interpret, the system uses rule-based analysis of size-stratified methylation patterns, which maintains predictive accuracy while producing results that are easier for clinicians to understand and interpret
3Reliability
If ctDNA is present at low levels relative to other DNA molecules, then disease detection sensitivity must be high, but reliability of extraction deteriorates
Solution Approach 1:
The patent applies local quality analysis by examining methylation patterns specifically within certain size ranges of cfDNA fragments, where tumor-derived DNA is more likely to be present. By focusing analysis on these specific local characteristics rather than treating all DNA equally, the system enhances its ability to detect low-level ctDNA signals with higher reliability
Data Source
AI summary
Methods for predicting a disease based on analyzing cfDNA fragments are described. The methods may comprise, for example, determining, by one or more processors, a test count of methylated loci for cfDNA fragments within a size range for a test sample obtained from a subject; determining, by the one or more processors, a disease score based on comparing the test count of the methylated loci to a reference count of methylated loci from one or more reference samples; and predicting, by the one or more processors, the disease state for the subject based on a comparison of the disease score to a predetermined disease score threshold.