Core SNP Selection Using Linkage Disequilibrium and Feature Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to effectively select core single-nucleotide polymorphisms (SNPs) with high heritability from biological SNPs, as only a few tens of thousands out of millions possess this characteristic.
Innovation Solution
A data processing method involving data dimension reduction and feature selection using linkage disequilibrium information and non-linear feature extraction methods to identify core SNPs with high heritability, employing techniques like filter, envelope, and embedded feature extraction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional SNP selection methods are used, then all biological SNPs can be analyzed, but the computational complexity and time consumption increase significantly
Solution Approach 1:
The patent segments the SNP selection process into three distinct stages: (1) initial filtering using linkage disequilibrium information to retain only independent SNPs, (2) dimensionality reduction using random projection to compress the feature space, and (3) final selection using feature extraction methods. This segmentation allows each stage to operate efficiently on reduced data, resolving the contradiction between comprehensive analysis and computational efficiency.
Solution Approach 2:
The patent performs preliminary filtering of SNPs based on linkage disequilibrium relationships before the main selection process. By pre-removing redundant SNPs that are in strong linkage disequilibrium, the system reduces the input data size for subsequent computationally intensive operations, thereby decreasing overall computational time while maintaining selection accuracy.
2Loss of information
If all biological SNPs are analyzed, then comprehensive genetic information is obtained, but data redundancy increases and processing efficiency decreases
Solution Approach 1:
The patent extracts and removes redundant SNP data through multiple mechanisms: (1) filtering out SNPs in strong linkage disequilibrium that provide redundant information, (2) applying random projection to extract only the most informative features, and (3) using feature extraction methods to identify core SNPs. This extraction process eliminates data redundancy while preserving essential genetic information, thereby improving processing efficiency.
Solution Approach 2:
The patent transforms the SNP data representation by changing parameters: (1) converting genotype data into linkage disequilibrium metrics, (2) applying random projection to transform high-dimensional data into a lower-dimensional space, and (3) using feature extraction to identify key parameters. These parameter changes reduce data complexity and redundancy while maintaining the essential genetic information needed for accurate SNP selection.
3Quantity of substance
If feature extraction methods are applied to reduce SNP data, then data dimension is reduced, but the complexity of the selection process increases
Solution Approach 1:
The patent segments the complex feature extraction process into three manageable stages with clearly defined functions: initial filtering based on linkage disequilibrium, dimensionality reduction via random projection, and final feature selection using extraction methods. This segmentation makes the overall complex process more tractable and easier to implement, resolving the contradiction between data reduction and process complexity.
Data Source
AI summary
A data processing method for selecting a core single nucleotide polymorphism (SNP) with a high heritability characteristic from biological SNPs includes obtaining a first SNP set; determining a second SNP set from the first SNP set based on first information of an SNP; and determining a third SNP set from the second SNP set based on a target feature extraction method and a non-linear feature extraction method. The first SNP set includes N SNPs, wherein N is a positive integer, the first information includes at least one of linkage disequilibrium information or a variance, the second SNP set includes M SNPs, wherein M is a positive integer, and M is less than N. The target feature extraction method includes at least one of a filter feature extraction method or an envelope feature extraction method.