Somatic Mutation Detection Using Neural Networks to Reduce NGS False Positives
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional mutation detection methods using next-generation sequencing (NGS) suffer from sequencing platform-specific false positives, which degrade the accuracy of mutation detection.
Innovation Solution
A neural network is trained to correct sequencing platform-specific false positives by generating and preprocessing genome data from target and normal tissues, and detecting mutations based on image data to improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional software is used to detect mutations through NGS, then mutation detection can be performed, but sequencing platform-specific false positives occur which degrade detection accuracy
Solution Approach 1:
The patent introduces an intermediary processing layer between raw NGS data and mutation detection results. This intermediary consists of multiple computational steps including quality score adjustment, position-specific error correction, and variant filtering that mediate the relationship between sequencing data and final mutation calls, thereby reducing platform-specific false positives while maintaining detection accuracy
Solution Approach 2:
The patent systematically changes multiple parameters throughout the data processing pipeline, including quality score transformations (Q-score adjustment), position-specific error probability modifications, and variant calling thresholds. These parameter changes are applied adaptively based on sequencing platform characteristics and genomic context to optimize both accuracy and reliability
2Productivity
If NGS is used instead of Sanger sequencing, then analysis time and cost are reduced, but sequencing platform-specific false positives increase
Solution Approach 1:
The patent applies preliminary error correction and quality adjustment steps before variant calling. By pre-adjusting quality scores and correcting position-specific errors in advance, the system prepares the data to minimize false positives during the high-speed NGS analysis, enabling fast processing without sacrificing reliability
Solution Approach 2:
The patent implements feedback mechanisms where error rates and quality metrics from sequencing data are continuously monitored and used to adjust processing parameters. This feedback loop allows the system to adapt to platform-specific error patterns in real-time, maintaining high productivity while correcting false positives through iterative parameter optimization
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
A mutation detection apparatus includes a memory configured to store software for implementing a neural network and a processor configured to detect a mutation by executing the software, wherein the processor is configured to generate first genome data extracted from a target tissue and second genome data extracted from a normal tissue, extract image data by preprocessing the first genome data and the second genome data, and detect a mutation of the target tissue on the basis of the image data through the neural network trained to correct a sequencing platform-specific false positive.