Device and method for screening key mutation influencing virus host adaptability based on deep learning model
By using deep learning models and convolutional neural networks to screen viral genome sequences and identify key mutations in viral host adaptation, this approach solves the problem of identifying the risk of cross-host transmission of viruses and screening mutations in existing technologies, and achieves efficient and low-cost assessment of viral host adaptation and mutation screening.
CN121905280APending Publication Date: 2026-04-21ACADEMY OF MILITARY MEDICAL SCIENCES
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ACADEMY OF MILITARY MEDICAL SCIENCES
- Filing Date
- 2025-12-23
- Publication Date
- 2026-04-21
Smart Images
Figure FT_1 
Figure FT_2 
Figure SMS_1
Abstract
The invention discloses a device and method for screening key mutations affecting virus host adaptability based on a deep learning model, and the method comprises the following steps: 1) based on a specified pathogen type, collecting gene sequence data of a pathogen and adopted host information, and preprocessing the collected data, constructing a model training data set and a test data set; 2) performing feature extraction on the collected sequence data to obtain a feature matrix of each sequence; 3) constructing a deep learning convolutional neural network model to perform data training and feature learning on the training data set; according to the method, the problem that measurement is inaccurate due to the fact that homology between virus hosts is high is solved, and the method has more accurate host adaptability prediction capacity and key mutation screening capacity.
Need to check novelty before this filing date? Find Prior Art
Citation Information
Patent Citations
Human endogenous retrovirus identification method and system based on deep learning
CN120526851A