A method for gene association analysis based on deep learning algorithm

Abstract

The invention discloses a method for gene association analysis on basis of a deep learning algorithm. The method based on SNP (single nucleotide polymorphism) set analysis needs to take SNP information, which comes from different positions of the same individual but is related, as a reference, and SNP of the individual is divided into multiple units according to existing biological knowledge. Firstly, all SNP is divided into multiple SNP sets on the whole chromosome level according to related knowledge of biology such as the principles approaching genomic characteristics. After division ends, each SNP set is input into an established bidirectional LSTM (long short-term memory) network, the network is a circulating neural network, and the state of the network contains old information of the last moment and also is a basis of weight change of the next moment. After LSTM network learning is completed, the attention degree required for input data can be output through network calculation. The method has better sensitivity and specificity and a new field is explored for development and research of the clinical medicine, genetic epidemiology and preventive medicine.

Description

Technical field [0001] The invention specifically relates to a gene association analysis method based on an LSTM (Long Short-Term Memory) network. The method is based on a deep convolutional neural network and a recurrent neural network model and belongs to the technical field of bioinformatics. Background technique [0002] The research on the association between gene chromosome base pairs and pathogenicity has always been one of the core research contents of bioinformatics. Carry out data mining in a huge database, deeply understand the complexity of organisms, and use existing knowledge and data to analyze as much as possible. However, due to the polymorphism of genes, there are often two or more discontinuous types in biological groups. Variant or genotype or alleles, so the use of machine learning methods with the characteristics of effectiveness and intelligence to study gene polymorphism can open up new research for the development of clinical medicine, genetics and preven...

AI Technical Summary

Problems solved by technology

However, the disadvantage of the traditional method is that it cannot handle variable-length sequences, and can only use fixed-length sequence fragments as input. Although a high prediction effect has been obtained, it was found in subsequent studies that the distance interval in the sequence is large. Residues can interact with each other, and researchers have to account for these errors

Images