Font Size: a A A

Study On Identification Of Functional Modules For Human Brain Specific Networks At Omics Scale

Posted on:2018-09-22Degree:MasterType:Thesis
Country:ChinaCandidate:W J CuiFull Text:PDF
GTID:2348330542479616Subject:Computer Science and Technology
Abstract/Summary:PDF Full Text Request
The human brain is the most complex structure in the known universe,it consists of billions of neurons and trillions of synaptic connections.The scientific research results that have been published have confirmed that brain was an organ that produces consciousness,thought and emotion.The human brain is composed of multiple organizations,these different parts in a complex network connect each other to coordinate human's behavior.Now the research results in fields of human brain genes and proteins mainly focused on mutation of a gene or effects of protease in human brain,which may lead to cognitive and other functional barriers.While in the aspect of the biology network conserved module discovery,many methods use the global network module identification algorithm.This paper mainly analyze the gene and protein networks of brain 14 tissues through dividing these networks into four development stage: embryo(prenatal),childhood(1 ~ 2 years old),adolescents(3 to 18 years),adults(18 years or older)and design a local network conserved module identification method to mine conserved modules in different tissues and different development stages.Then we try to analyze relationship between human brain function and human brain development from the perspective of molecular biology.This paper firstly uses the shortest path method to construct strongly associated protein networks and the correlation method to construct specific gene networks.Then the reliability of these networks constructed are verified by normalized entropy.For problems of identification of conserved modules,we design a method which integrates GO semantic similarity and protein homology to to evaluate the conservativeness of the modules.This method is different from many existing algorithms which use the discovery of global network conserved method and focus on local networks.First,function modules are mined through other module mining algorithms,then conservative score of two modules is calculated by judging the degree of conservation between the modules.Then horizontal comparison is executed when the development stage is the same and tissues are different,longitudinal comparison is executed when the tissue is the same and the development is different.The experimental results show that,for those conserved modules in different development stages,if they are responsible for basic functions,its corresponding conserved module function usually does not change in the development process.If they are responsible for non-base function,these modules reflect the specificity associated with the developmental stage.These modules which possess developmental specificity may be new targets for curing age-related diseases.
Keywords/Search Tags:Conserved module, Local network alignment, Multiple developmental stages, Human brain tissue, Protein network
PDF Full Text Request
Related items