Microarray Data Mining And Bioinformatic Analysis Of Middle Cerebral Artery Occlusion

Part I:BackgroundCerebrovascular disease is a threat to human life and health around the world. It ranks the second on the list of diseases causing human deaths after cancer in the world. At present, about 150-180 million new cases occur per each year in China, with nearly one million deaths. It is the main cause of death in the city in China. According to the third national death review sample survey, cerebrovascular disease has risen to the first reason of the death for our country. Among the cerebrovascular diseases, ischemic cerebrovascular disease is the most common, the incidence rate of which accounts for 70%-80% in all strokes. Intracranial stenosis is an important cause of ischemic cerebrovascular disease, among which, middle cerebral artery stenosis or occlusion has the highest incidence rate, accounting for 66%-73.3% in intracranial arterial disease. Study has found that the recurrence probability of ipsilateral cerebral ischemic events in the person with middle cerebral artery occlusive disease (MCAOD) is about 9.1%, besides, the Chinese people have a higher chance of recurrent cerebrovascular events. Therefore, it is necessary to study the development of MCAOD at the molecular level, which may play important roles in the prevention and treatment of this disease.Gene chip is an important technology platform in the life sciences in the 21st century. It is an effective means of screening the differentially expressed genes, with advantages of high throughput and rapid measurement. Due to the advantages of microarray on investigating gene expression model, microarray has been widely used in disease pathogenesis, early diagnosis, treatment and prognosis and other research areas. With the development of microarray technology, a massive complex biological data appeared. How to interpret the hybridization information of thousands of genes points on the chip and then to reveal their characteristics and laws has become a major problem that restricts the application and development of gene chip technology. Bioinformatics is an interdisciplinary subject that integrates the knowledge of information, statistics and computer science to analyze the biological information contained in massive biological data. By filtering massive data of biological chips and utilizing the techniques such as sequence alignment, statistical analysis, clustering, visualization, and biological pathway analysis to implement data mining, bioinformatics enriches the understandings about disease occurrence and progression from molecular level. With the development of bioinformatics, a new biological research pattern is emerged. In this research pattern, researchers propose hypothesis based on the existing data and then test the hypothesis.This research aimed to screen the differentially expressed genes associated with MCAOD based on the GEO database and two gene expression profile datasets. Additionally, we analyzed the interaction between these differentially expressed genes and the enrichment functions of these differentially expressed genes by combining with bioinformatics software and text mining techniques, so as to explores genes, pathways, transcription factors, and miRNAs associated with MCAOD. It would provide critical information for better understanding on the molecular mechanism of MCAOD occurrence and development.Part ?:Bioinformatics analysis of differentially expressed genes in middle cerebral artery occlusion (MCAO)Objective:The biological processes of MCAO are very complicated, therefore, it becomes an important way to screen the differentially expressed genes associated with this disease from genomic level. In this chapter, we screened the differentially expressed genes associated with MACO-induced ischemic stroke based on the microarray data of GSE35338. The obtained differentially expressed genes were then performed functional annotation, pathway enrichment analysis and protein-protein interaction network analysis, which may provide a theoretical basis for exploring the molecular mechanism of MACO.Methodology:In this research, we downloaded the microarray dataset of GSE35338 from Gene Expression Omnibus (GEO) database, which is based on the Affymetrix Mouse Genome 430 2.0 Array. A total of 21 specimens, including one day (n= 5), three days (n= 3) and seven days (n= 3) after middle cerebral artery occlusion (MCAO) produced ischemic stroke as well as one day (n= 4), three days (n= 3) and seven days (n= 3) after control sham surgeries specimens, were available based on the GPL 1261 Platform. After data preprocessing, the differentially expressed genes (DEGs) between MCAO-induced ischemic stroke and sham controls at various time points were identified with the cutoff values of |logFC|> 1.5 and p-value< 0.05. Then we used analysis tools in DAVID software for GO analysis and KEGG pathway enrichment analysis. The differentially expressed genes were then performed Venn Diagram analysis and the overlapping genes were identified. The identified differentially expressed genes were then imported into STRING online database for protein-protein interaction network analysis, and computed the network topology through Cytoscape software. Finally, we identified network clusters in these networks by using the MCODE plugin.Results:We selected 294 genes as differentially expressed genes between 1 d MCAO and sham specimens,87 differentially expressed genes between 3 d MCAO and sham samples and 57 differentially expressed genes between 7 d MCAO and sham controls. All the enriched genes were up-regulated in MCAO. The differentially expressed genes for 1 d MACO were mostly located in response to wounding and immune response. The differentially expressed genes for 3 d MACO were mainly enriched in extracellular region. The enriched GO terms of differentially expressed genes for 7 d MACO were related to immune response, and inflammatory response. A total of 337 differentially expressed genes were obtained after Venn Diagram analysis. In the protein-protein interaction network,22 nodes were screened with degrees larger than 10, thereinto, the degrees of Cxcl10 and 116 were more than 20. Five sub-networks were obtained using MCODE, the functions of which were mainly related to the cell cycle, immune response, response to wounding and cell proliferation.Conclusion:(1) The differentially expressed genes were performed GO analysis and KEGG pathway enrichment analysis, which may provide theoretical base for the lab research on MACO. (2) We constructed protein-protein interaction network of differentially expressed genes. Cxc110 and 116 may be involved in the development of MACO, which may provide novel direction for the diagnosis and treatment of this disease.Part III:Study on molecular mechanism of the role of PACAP38 in middle cerebral artery occlusion (MCAO)Objective:Three hours after ischemia was the "best time" of reperfusion. It is of vital importance to give effective and timely treatment for cerebral ischemia patients within 6 hours or longer period of time. In this chapter, we screened the differentially expressed genes associated with MACO after PACAP38 treatment based on the microarray data of GSE62884. The obtained differentially expressed genes were then performed functional annotation, pathway enrichment analysis and protein-protein interaction network analysis, which may provide a theoretical basis for understanding the molecular mechanism of PACAP38 on MACO.Methodology:In this research, we downloaded the microarray dataset of GSE62884 from Gene Expression Omnibus (GEO) database, which was a dual-channel chip, containing eight samples:ischemic core regions at six or 24 h post-PACA38 or saline injections and penumbra regions at six or 24 h post-PACA38 or saline injections. After data preprocessing, the differentially expressed genes (DEGs) were identified with the cutoff values of |logFC|> 0.58 and p-value< 0.05. Then we used analysis tools in DAVID software for GO analysis and K.EGG pathway enrichment analysis. The identified differentially expressed genes were then imported into STRING online database for protein-protein interaction network analysis, and computed the network topology through Cytoscape software. The network clusters in the network were identified by using the ClusterONE. The differentially expressed genes were then performed Venn Diagram analysis and the overlapping genes were identified and analyzed for transcription factor prediction.Results:A total of 123 differentially expressed genes were identified in the group of ischemic core (6 h); 2009 in the group of ischemic core (24 h); 126 in the group of penumbra (6 h); 1484 in the group of penumbra (24 h). The differentially expressed genes in ischemic core group were mostly located in chemokine activity and immune response. The differentially expressed genes in penumbra group were mainly enriched in cell proliferation, immune response and extracellular region. The protein-protein interaction network of ischemic core (6 h) group comprised 16 nodes and 14 interaction pairs, of ischemic core (24 h) group comprised 961 nodes and 4633 interaction pairs which had four significant sub-network modules. The protein-protein interaction network of penumbra (6 h) group comprised 33 nodes and 30 interaction pairs, of penumbra (24 h) group comprised 681 nodes and 2873 interaction pairs which had three significant sub-network modules. A total of 1039 differentially expressed genes were obtained after Venn Diagram analysis. Cxcl10 and Il1b were up-regulated in two tissues and two time points. The encoding genes of transcription factors Irf1, Bcl3, Nfkb1, Relb, Hic1, and Ets1 were also differentially expressed genes.Conclusion:(1) The differentially expressed intersection genes were performed GO analysis and KEGG pathway enrichment analysis, which provided theoretical base for the lab research on MACO. (2) We constructed protein-protein interaction network of intersection genes and found that Cxcl10 and Il1b were up-regulated in two tissues and two time points, which may play an important role in the treatment of PACAP38 on MACO.Part ?:Integrated analysis of microarray GSE35338 and GSE62884Objective:The intersection genes of the two datasets were identified, and were performed functional enrichment analysis, protein-protein interaction network analysis, and transcription factors prediction. We aimed to further explore the genes or pathways which may play important roles in the occurrence and development of MACO.Methodology:The microarray of GSE35338 identified 337 differentially expressed genes and GSE35338 identified 1039 differentially expressed genes. Based on the genes in two dataset, we performed intersection analysis. The intersection genes were performed GO and KEGG pathway enrichment analyses using DAVID online tool. Then the correlation network were constructed based on the identified GO terms and KEGG pathways using Enrichmentmap plugin of Cytoscape. The intersection genes were then imported into GeneMANIA for function network analysis. The transcription factor and miRNA were identified.Results:A total of 103 intersection genes were identified based on the microarrays of GSE35338 and GSE62884. These intersection genes were significantly enriched in the GO terms and pathways associated with immune response, response to wounding and inflammatory response. After being analyzed by Enrichmentmap, GO:0009611-RESPONSE TO WOUNDING, GO:0006954-INFLAMMATORY RESPONSE, MMU04620:TOLL-LIKE RECEPTOR SIGNALING PATHWAY, GO:0005576-EXTRACELLULAR REGION were found to have the highest degrees. The function network included 97 nodes and 97 functional relationship pairs. The transcriptional regulatory network included 75 transcription factors, among which Bcl3, Irfl, Maff, Fosll, Stat6 and Stat3 had higher degrees. Additionally,2635 miRNA-target genes interaction pairs were obtained including 668 miRNAs, such as miR-124 and miR-29c.Conclusion:(1) The intersection genes that were differentially expressed in two tissues were performed GO analysis and KEGG pathway enrichment analysis, and they were significantly enriched in the functions associated with immune response, response to wounding and inflammatory response. (2) We constructed function network, and transcriptional regulatory network, and predicted the miRNA-target genes interaction pairs. These transcription factors and miRNAs may play important roles in the occurrence and development of MACO.