| Objectiv: To investigate the differentially expression level of both white blood cells(WBC)miRNAs and plasma miRNAs in Schizophrenia patients.Explore fast,sensitively and specificly biomarkers,made diagnostic model to provide laboratory evidence for the diagnosis,disease assessment,treatment,prognosis and so on.Methods: 1 The screening and validation of differential expression microRNAs: eight significant correlation microRNAs in schizophrenia were selected as target microRNAs.90 Schizophrenia(SZ)cases,90 bipolar disorder(BPAD)cases,and 90 healthy control group cases were selected as the research object,collected their whole blood,extract out WBC and plasma,the qRT-PCR was used to detect the expression level of eight microRNAs in the two types samples.the SPSS20.0 software was used to analyze the expression level of target microRNAs,verify weather the target miRNAs have a significant differentially expression.2 ROC curve: Schizophrenia as case group,non schizophrenia as the control group,the differential expression of miRNAs were made the ROC curve.3 Diagnosis models: 3 types diagnosis model establishment: 45 cases of schizophrenia,90 cases of no schizophrenia(45 cases bipolar disorder and 45 cases Health people)as a training sample,and the rest cases as a validation sample.The decision tree,artificial neural network and logistic regression analysis were established in plasma and WBC.Results: 1 RT-PCR validation results: the three groups of samples were analyzed by the non parametric Kruskal-Wallis test in the SPSS 20 software and compared with the other two groups.In plasma samples,miRNAs in SZ compared with both BPAD and healthy group,miR-31-5p P>0.001,the difference was not statistically significant,the seven other miRNAs: miR-99b-5p,miR-107,miR-134-5p,miR-487b-3p,miR-181b-3p,miR-431-5p and miR-433-5p was expression decreased,the difference was statistically significant.The WBC's miRNAs in SZ compared with BPAD,miR-181b-3p has no statistically significant difference.The other seven miRNAs have significant difference expression;in SZ compared with healthy control group,miR-433-5p's P>0.001,the difference was not statistically significant,and the other seven miRNAs had statistical significance.2 ROC curve : In Schizophrenia as disease group and no schizophrenia as control group,there was statistical significance of the 13 indicators been seleted to made the ROC,and the AUC were: plasma miR-99b-5p(0.923),miR-181b-3p(0.902),miR-107(0.744),miR-433-5p(0.741),miR-431-5p(0.731),miR-134-5p(0.715)and miR-487b-5p(0.689);WBC miR-31-5p(0.931),miR-487b-3p(0.887),miR-99b-5p(0.869),miR-431-3p(0.763),miR-134-5p(0.756)and miR-107(0.721).4 diagnosis models: 1)Decision Tree model: the plasma miRNAs' s optimal model 5,validation sample accuracy of plasma 85.2%,sensitivity of 97.8% and specificity of 78.9%;WBC optimal model 5,validation sample accuracy 88.1%,sensitivity 77.8%,specificity 93.3%.2)ANN model: the plasma optimal model 2,validation sample accuracy rate of 91.9%,sensitivity 97.8%,specificity 88.9%,WBC optimal model 6,the accuracy is 92.6%,sensitivity 86.7%,specificity 95.6%.3)Binary Logistic Analysis model: plasma optimal model 2,validation sample accuracy rate is 90.4%,sensitivity 95.6%,specificity 87.8%.The accuracy of the model was 88.9%,the sensitivity was 80%,and the specificity was 93.3%.Conclusion: 1.The Schizophrenia's plasma miR-107,miR-433-5p,miR-99b-5p,miR-134-5p,hmiR-487b-3p and miR-431-5p showed lower expression;miR-181b-3p significantly higher expression;white blood cell miR-31-5p,miR-107,miR-99b-5p,miR-134-5p,miR-487b-3p and miR-431-5p significantly lower expression,all of these miRNAs can be to Schizophrenia's biomarker.2.plasma miR-99b-5p,WBC miR-134-5p and miR-433-5p have an significant difference expression in BPAD.3.establishment of diagnostic model can improve the diagnostic value of miRNAs in schizophrenia.The diagnostic value of miRNAs in white blood cell sample is higher than that in plasma miRNAs. |
PDF Full Text Request