Genetic Testing Using Feature Generation Network for Low-Depth Data

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The high cost of genetic testing due to the requirement for deep sequencing, which is expensive and not widely accessible, necessitates a method to reduce the cost and resource intensity while maintaining accuracy.

Innovation Solution

A genetic testing method and system that processes low-depth genetic data using a feature generation network layer for feature extraction and a genetic identification network layer for testing, generating a model that performs accurately on low-depth genetic data, thereby reducing processing resources and costs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If deep sequencing is used to ensure accurate genetic testing results, then measurement precision is improved, but cost and resource consumption increase significantly

Engineering Contradiction:

Improvegenetic testing accuracyVSAvoiddata processing resources

Core Design Contradiction:

Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent extracts and processes only the most relevant genetic features from low-depth sequencing data using a feature generation network layer, rather than processing all genetic data at full depth. This selective extraction maintains testing accuracy while significantly reducing the quantity of data that requires expensive deep sequencing processing

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent transforms the problem from a single dimension (sequencing depth) to multiple dimensions by introducing feature extraction and enhancement layers. Instead of relying solely on increasing sequencing depth, the system processes low-depth data through multiple network layers that extract, enhance, and reconstruct genetic features, achieving accurate results without proportionally increasing data volume or cost

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

2Reliability

If deep sequencing is performed to maintain testing accuracy, then reliability is improved, but cost increases

Engineering Contradiction:

Improvegenetic testing reliabilityVSAvoidcost of genetic testing

Core Design Contradiction:

ReliabilityVSEase of manufacture

Solution Approach 1:

The patent changes the parameter of sequencing depth from high to low, and compensates for the reduced depth through parameter optimization in the neural network processing stage. The feature generation network layer and genetic identification network layer are optimized to extract maximum reliability from low-depth data, achieving comparable testing reliability at lower cost

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent introduces an intermediary feature enhancement process between raw low-depth genetic data and final testing results. The feature generation network layer acts as an intermediary that processes, enhances, and reconstructs genetic features, bridging the gap between low-depth input data and high-reliability testing outcomes without requiring expensive deep sequencing

Inventive Principle:

Principle #24Intermediary (Mediator)

3Quantity of substance

If low-depth genetic data is processed directly without feature enhancement, then processing resources are reduced, but measurement precision deteriorates

Engineering Contradiction:

Improvedata processing resourcesVSAvoidtesting accuracy

Core Design Contradiction:

Quantity of substanceVSMeasurement precision

Solution Approach 1:

The patent performs preliminary feature extraction and enhancement on low-depth genetic data before final testing analysis. The feature generation network layer pre-processes the low-depth data to extract and enhance critical genetic features, preparing optimized input for the genetic identification network layer. This preliminary action ensures that subsequent testing achieves high precision without requiring intensive processing of raw low-depth data

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent replaces the mechanical approach of increasing sequencing depth with a computational approach using neural network-based feature extraction and enhancement. Instead of physically obtaining more sequencing reads through deeper sequencing, the system uses the feature generation network layer to computationally extract and enhance genetic features from low-depth data, achieving high precision through intelligent processing rather than brute-force data accumulation

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS20220398435A1Genetic Testing Method, Model Training Method, Apparatus, Device, and System

Publication Date: 2022.12.15 ALIBABA INNOVATION PRIVATE LIMITED

AI summary

Methods, apparatuses, devices and systems for genetic testing and model training are provided. A genetic testing method includes: obtaining genetic data to be processed, an average number of genetic segments corresponding to each position in the genetic data to be processed being less than or equal to a preset threshold; inputting the genetic data to be processed into a feature generation network layer for performing a feature extraction operation to obtain genetic features corresponding to the genetic data to be processed and enhanced features corresponding to the genetic features; and inputting the genetic data to be processed and the enhanced features into a genetic identification network layer for performing a genetic testing operation to obtain a testing result. The present disclosure realizes performing feature extraction operations through low-depth genetic data, obtaining genetic features and enhanced features corresponding to the genetic features, and performing testing operations based on the enhanced features.