Genome big data-oriented GWAS-AI sequence depth fusion prediction method
By employing the GWAS-AI sequence deep fusion prediction method, which combines GWAS statistical weights, individual genotypes, and AI sequence embedding vectors to construct a deep fusion feature matrix, the problem of inaccurate phenotypic prediction in existing technologies is solved, and high-precision phenotypic prediction is achieved.
CN121687203APending Publication Date: 2026-03-17CHINA AGRI UNIV
View PDF 0 Cites 0 Cited by
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-10
- Publication Date
- 2026-03-17
Smart Images
Figure CN121687203A_ABST
Abstract
The invention belongs to the technical field of genome data processing, bioinformatics and artificial intelligence crossing, and particularly relates to a genome big data-oriented GWAS-AI sequence depth fusion prediction method. According to the method, deep multiplication or weighted interaction is carried out on GWAS statistical significance, AI sequence function embedding and individual genotypes instead of simple feature splicing for the first time, so that the model can learn how specific function sequences (AI embedding) on important sites (GWAS weight) are expressed in specific individuals (genotypes); and the biological significance and predictive ability of the features are greatly enhanced. The invention provides a repeatable end-to-end automatic workflow from original VCF / phenotype data to a final prediction model and a visual report, and is suitable for genome big data processing. Through deep fusion features and optimized AI model training, a complex nonlinear genetic effect can be captured, and a high-precision prediction result is obtained.
Need to check novelty before this filing date? Find Prior Art