Research Of Genes In Molecular Mechanism Of Complex Diseases

Prevalence and development of complex diseases like cancer, metabolic diseases, heart andcardiovascular diseases, not only bring human beings a long period of pain and suffering, but alsobe a heavy burden to the world. Thus, it becomes a most pressing medical problem to study thepathogenesis of complex diseases more extensively and deeply. However, the diagnosis ofcomplex diseases are different from those of traditional single-gene genetic diseases, asenvironmental factors and genetic factors are needed to be considered to provide solid foundationand breakthrough ideas for diagnosis, drug design, personalized treatment and prevention.Transcriptome data plays an important role in revealing disease-related genes and molecularmechanism. In recent years, some researches based on time series expression data reveal muchcrucial targets in the process of cancer development, which provide meaningful insights in furtherstudy on regulation of gene expression, metabolic pathways and signal transduction. However, it’sdifficult to analyze these hidden changes of gene expression with commonly used statisticalmethods, so we created a complete set of expression profiling data analysis process used to screenout those core regulatory genes closely related with the disease to describe its specific molecularmechanisms in detail on the basis of existing algorithms. Conservative and dynamic cooperativeproperties were firstly introduced to mine gene expression profiling data for finding functionalgene signatures. TiCoGE is just a strategy to find those gene clusters with strong cooperativedynamic conservative properties across cancer progression stages in dealing with gene expressionprofiling data, and each gene will be ranked with a conservative score to evaluate its importance incancer progression and development. Our methodology was presented in an effort to breakthrough the limitations of pure statistical models and looks reasonable sensitive for further studyof gene function and regulation of networks.We chose a set of time series gene expression data on prostate cancer, including the fivestages for analyzing the co-expression of genes in the process of tumor development. TiCoGEmethod eventually screened out492high score genes, and several genes were randomly selectedand all annotated to well known oncogenesis mechanisms or disease models, such as humanembryonal cancer stem cells, proliferative disorders, human immunodeficiency, chronicinflammation, out of control of remodeling balance, DNA-damaging, etc. These results not onlyconfirmed the reliability and validity of TiCoGE method, but also explore the relationship ofmultiple genes in cancer development from a global perspective. It is valuable to elucidate thelisted core genes for further research as an analytic basis and data accumulation. Our work also included analyzing whole expressed miRNA of three artificial pluripotent cells,by an innovative use of focusing on the region of differentially expressed miRNAs, the adjacentmiRNA target genes and imprinted genes. Besides, another part of our work is about the molecularmechanism of anti-diabetic drugs, including rosiglitazone, pioglitazone and metformin. in fivedifferent tissue cells, that is liver, heart and3adipose tissues. We explained the potential same ordifferent mechanism of drugs and search the diverse function and differentiation between threeadipose tissues, which may provide new idea for anti-diabetic drug design.Thus, to some extent, our evolutionary conservation analyzing based methodologycompensates for the inherent weaknesses of current statistical methods and provides a new wayfor gene expression profile analysis. However, complex disease is caused by multiple genesthrough protein complexes, regulatory networks, as well as the interaction path network control,we need to use multiple types of data and tools for integration analysis in order to clarify theassociation between the combination of multiple genes and complex diseases gradually.