Study On Global Human Mitochondrial DNA Development Tree Based On Integrated Learning

In the late twentieth Century,with the in-depth study of the origin of the problem,the mtDNA problem has gradually entered the field of vision of scholars.There are three main characteristics of mtDNA:the strict maternal inheritance,lack of recombination,and the high degree of population variation.Based on this feature,mtDNA provides a good genetic marker for researchers,which provides a basis for the study of human genetic relationships and genetic differentiation within populations,further promoting this research.After entering in twenty-first Century,the researchers launched an analysis of mtDNA sequences of different nations in different countries.Based on this feature,mtDNA provides a good genetic marker for researchers,which provides a basis for the study of human genetic relationships and genetic differentiation within populations,further promoting this research.The main work of this paper is to analyze the classification of individual mitochondrial DNA information in the global human mitochondrial DNA phylogenetic tree.It is also the innovation of this paper to complete the integrated learning algorithm.Aim at the problem that haplogrep,a mainstream software at home and abroad,only supports batch learning,cannot update the standard data set,and the lower accuracy,this paper presents an algorithm based on integrated learning.The main work of this paper is:Innovatively proposes an application algorithm based on integrated learning,using the tensorflow framework and implemented in python language.Tests have shown that the correct rate can be increased by about 18%relative to the mainstream software algorithm.(1)For the first time using an incremental learning algorithm,localized data can be entered into the model.Whenever there is new data updated,the model weight can be self-updated and then a localized model is established.It is further confirmed by experiments that this will allow localized data testing to be more accurate.(2)The paper uses the integrated learning algorithm,built a classify model having the sparse data analysis ability of naive Bayesian algorithm and the powerful abstract expression ability of the neural network.(3)The target of this algorithm is the global human mitochondrial DNA phylogenetic tree spectrum,genealogy in the data update,we can automatically update the localization model.The results show that the proposed algorithm can improve the performance of the traditional program and provide reference for the design and implementation of other algorithms in bioinformatics.The results show that the proposed algorithm can improve the performance of the traditional program and provide reference for the design and implementation of other algorithms in bioinformatics.