Computational protocol: Serum and Extracellular Vesicle MicroRNAs miR-423, miR-199, and miR-93* As Biomarkers for Acute Graft-versus-Host Disease

Similar protocols

Protocol publication

[…] MicroRNA expression was assessed using SDS2.4 software (LT). Input data were based on converted ΔCt values, according to the comparative Ct method (). Statistical analyses were carried out using SPSS v22.0, Sigmaplot v12.5, or GraphPad Prism v6.0. Graphs were produced using Sigmaplot v12.5 or GraphPad Prism v6.0. Differences between groups were assessed using the Student’s t-test (two groups) or one-way ANOVA (less than two groups). Homogeneity of variance was assessed using the Levene statistic. Following a significant ANOVA result, differences between pairs of groups were assessed via the Tukey post hoc test (SPSS v22). Receiver operating characteristic (ROC) analysis was performed using disease status as the binary state (classification) variable and marker expression on a continuous scale as the test variable (SigmaPlot v12.5). The ROC post-test results used a pre-test prior-probability of 0.5 and cost ratio of 1.0 (). The optimal cutoff value to dichotomize miRNA expression was computed from sensitivity and specificity using the slope m by finding the cutoff that maximizes the function: sensitivity − m (1 − specificity) (SigmaPlot v12.5) (). The accuracy of the test was defined by the area under the curve (AUC) whereby AUC = 0.5 means no diagnostic ability and AUC = 1 means perfect diagnostic ability. Principle components (PCs) were calculated for correlated miRNAs using JMP® Pro v11.2. Correlation between miRNA expression was determined using Pearson’s correlation with Holm–Bonferroni multiple comparisons adjustment applied (). The first PC1 in all analyses explained the majority of the variance (range 63.5–78.2%) and thus, was used for composite ROC analysis. The weighting used in PC1 for each miRNA was derived from eigenvectors of the correlation matrix. PC1 took the form: “Component 1 = (weighting1 * StdmiR1) + (weighting2 * StdmiR2) + …….” where “StdmiRX = ΔCtmiRx–mean(ΔCtmiRx)/SD (ΔCtmiRx)” (JMP® Pro v11.2). Survival plots were generated using the Kaplan–Meier method and differences in outcome were assessed for significance using the log-rank test (SPSS v22). Cumulative incidence based on the competing risk method, as described by Fine and Gray (), was used for assessing the association between miRNAs and both relapse and non-relapse mortality () using R (R Project) package “cmprsk” (competing risks) (). […]

Pipeline specifications

Software tools SigmaPlot, JMP Pro
Application Miscellaneous
Diseases Graft vs Host Disease