| description |
MicroRNAs (miRNAs) are short, non-coding RNAs, whichpost-transcriptionally regulate gene expression and are proposed to play akey role in the regulation of biological process and homeostasis. miRNAexpression has been widely analyzed by microarray, however, nextgeneration sequencing (NGS) analysis has higher reproducibility andsensitivity than microarray analysis. Here, we report a comparison miRNAexpression profiling in hepatocellular carcinoma (HCC) examined by NGS andmicroarray. Materials and methods:Total RNA (14 HCCs and 6 surrounded non-tumor tissues) was converted intocDNA libraries with adapter sequences. Barcoded templates were sequencedwith Illumina MiSeq 50-bp paired-end reads. miRNA expression profile wasestablished by NGS. The result of sequencing data was analyzed by usingmiRdeep2. miRNA expression profile was in 10 from 14 HCCs was alsoestablished by microarray (Agilent human microRNA microarray Rel 14.0) andmicroarray data was analyzed by GeneSpring GX10.0.2 (Agilent).Results:The average of total sequencing reads was over 2.2 million and thesequencing reads that were mapped on human genome sequence, wasapproximately 57%. Using 10 HCC templates, the correlation of miRNAsexpressions between NGS and microarray was examined. The averagecorrelation logarithmic index of miRNA expressions was 0.613. Thesubtraction of correlation logarithmic index in miRNA expression level was0.587. As a result of investigating the same sample plural times, thereproducibility miRNA expressions was 0.976. The expression level of 8miRNAs (miR-10a-5p, miR-122-5p, miR-146b-5p, miR-148a-3p, miR-192-5p,miR-22-3p, miR-26a-5p, and miR-27b-3p) in non-tumor tissue was higher thanthat in HCC (p<0.05). The expression level of 3 miRNAs (miR-10b-5p,miR-143-3p, miR-21-5p) in non-tumor tissue was lower than that in HCC(p<0.05). The diagnostic accuracy, p-value, and AUC in HCC by using these11 miRNAs expression pattern are 90.0%, 7.224E-04, and 0.92 of AUC,respectively. Finally candidates of novel miRNAs were also detected by NGSanalysis.Conclusions:The reproducibly of the miRNA expression profile using NGS analysis wassatisfied and miRNA expression profile by using NGS was concordance with that by microarray. Moreover, the discovery of the novel miRNAs canbe easily applied by NGS. |