Transcriptome profiling of prostate tumor and matched normal samples by RNA-Seq.

BACKGROUND: RNA-Sequencing (RNA-Seq) has greatly influenced cancer researches, and it provides an unprecedented resolution in estimating gene expression and has less signal noises compared to cDNA microarray. AIM: We aimed to identify a list of protein-coding genes and lincRNAs that are expressed differentially between tumor and normal tissues. MATERIALS AND METHODS: In this study, we analyzed including 10 human prostate tumor tissues and their matched normal tissues transcriptome dataset generated by recently developed RNA-Seq technology. RESULTS: By aligning short reads to human RefSeq genes and lincRNAs, we identified 10 RefSeq genes that were differentially expressed between tumor and normal samples with a p-value < 0.05, which were sufficiently enough to distinguish these two groups. Further loosing the p-value cutoff to 0.1 identified an lincRNA which is antisense to Cullin-associated and neddulation-dissociated 1 (CAND1), whose expression is repressed in prostate tumor cells. By examining the expression of CAND1 and its antisense lincRNA in the transcriptome dataset, we found an interaction between them as high expression of CAND1 and low expression of lincRNA is normal samples, and verse visa in tumor samples. CONCLUSIONS: These findings suggest the important usage of RNA-Seq in cancer studies for biomarker development and functional investigation.