DNA Methylation Measurement Using Standard Deviation Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining DNA methylation levels across multiple loci and samples lack precision in distinguishing between statistically significant differences, as the error associated with measurements is often unknown, making it difficult to determine if methylation levels between cell populations are truly different.
Innovation Solution
A method and system that utilize a correlation between methylation levels and their standard deviations, where data representing standard deviation values are correlated with methylation levels, allowing for the comparison of methylation levels between samples by referencing a training set that includes mixtures of DNA with varying methylation levels, enabling the determination of whether methylation levels are the same or different based on standard deviations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If methylation levels are measured across multiple loci and samples, then comprehensive methylation data is obtained, but precision in distinguishing statistically significant differences deteriorates due to unknown measurement errors
Solution Approach 1:
The patent applies preliminary action by creating a training set with known methylation levels and predetermined standard deviations before analyzing test samples. This training set is used to establish correlation data between methylation levels and standard deviations in advance, which then serves as a reference for determining statistical significance in subsequent measurements without requiring extensive recalculations.
Solution Approach 2:
The patent introduces an intermediary approach by using standard deviation values as a mediating parameter between raw methylation measurements and statistical significance determination. Instead of directly comparing methylation levels, the system uses standard deviations (derived from the training set) as an intermediary to assess whether observed differences are statistically significant, thereby resolving the precision problem.
2Measurement precision
If standard statistical analysis is performed on each test sample, then accurate methylation level comparison is achieved, but extensive recalculations increase time consumption
Solution Approach 1:
The patent performs preliminary statistical analysis by predetermined standard deviations for various methylation levels using a training set before actual testing. This advance preparation creates a lookup reference that eliminates the need for extensive recalculations during test sample analysis, significantly reducing time consumption while maintaining comparison accuracy.
Solution Approach 2:
The patent uses copying by creating a training set that replicates the statistical characteristics of the measurement system. The standard deviations determined from this training set are copied and applied as reference values for evaluating test samples, avoiding redundant statistical computations while preserving the accuracy of significance determination.
3Ease of operation
If methylation levels are determined without reference to standard deviation data, then measurement process is simplified, but ability to determine statistical similarity deteriorates
Solution Approach 1:
The patent uses standard deviation data as an intermediary that bridges the gap between simple measurement processes and rigorous statistical determination. The standard deviations serve as a reference framework that can be applied without complicating the actual measurement process, allowing researchers to maintain operational simplicity while achieving accurate statistical similarity determination through comparison against predetermined values.
Data Source
AI summary
The present embodiments relate to a system and method of measuring the methylation level of DNA. Some embodiments relate to a system and method of measuring methylation level of DNA with a gene array.