| Objective:Acute myeloid leukemia(AML)is one of the common malignant diseases of the blood system,the prognosis varies in the same type of AML.There are many factors which may influence the prognosis of the disease.Based on cytogenetic and combine different types of gene mutation,we can divide AML into different risk stratification groups as ELN and NCCN guidelines suggested.People with high risk AML often presents poor outcome and more rapidly disease progression.So it is necessary to determine the risk stratification of AML during the first diagnose and adjust treatment accordingly.In this study,we use information downloading from TCGA database to analyze the differential expression of mRNA between low and high risk stratification leukemia,and study the differential genes by bioinformatics methods,during which,we hope to find out mRNA that associated with disease prognosis and provide a biomarker for further study.Methods:This study use the mRNA and clinical data downloading from the TCGA database,then divide the clinical data into different groups by risk stratification.Then compare different groups looking for differential expression of genes.Screening the genes by certain criteria,then performing GO analysis to understand the function of the different genes.Then do a KEGG analysis to see if the genes belong to some specific pathways.Use STRING database to form a network between proteins.Import the network into Cytoscape,and calculate the topology parameters of the network,select hub gene with degree>10.By using MCODE plug-in of Cytoscape to find modules in the network.Then select a module with specific hub gene,and export the genes the module contained for further study.Perform survival analysis to determine genes in the select module that may have a connection with prognosis.And then a clinical association analysis to identify the genes associate with clinical feature.At last,by using the collected clinical samples to preform RT-PCR to testify the genes I selected are in consistence with the database.At last,analyze the relationship between the selected genes and clinical indicators.Results:With bioinformatics analysis above,this study found out 512 genes that are significant between low and high risk stratification of AML.And these genes can be concluded into 13 pathways.By analyzing the interaction between the proteins,we found 24 hub genes and 13 sub-network in the whole PPI network.In the sub-network I choose to continue on studying contains 7 genes.After survival analysis,there remain 2 genes that is significant to AML survival.At last,I choose ALDH1A1 gene to carry on further study.The expression level of gene ALDH1A1 in TCGA database is in consistence with relative expression level of samples I collected in clinic,which have a significant difference between low risk group and high risk group(P<0.05).And there is a difference between age group(60 is the bar).On the contrast,there is no significant meaning of ALDH1A1 relative expression in different age groups(65 is the bar),different gender groups and different FAB subgroups(M2 and M5)(P>0.05).Conclusion:This study found that there were many differential genes(p<0.01,FC absolute value>2)between the low-risk group and the high-risk group of AML,and a significant proportion of the genes were associated with prognosis.PPI network between the differential genes are successfully constructed.By performing RT-PCR in clinical samples we also verified that the relative expression of hub gene ALDH1A1 is different between low and high risk group which is in consistence with what we analyzed from TCGA,and that may provide a new gene locus for detect in the future. |
PDF Full Text Request