College of Bioinformatics Science and Technology, Cancer Hospital Affiliated to Harbin Medical University, Harbin, 150081, China.
Nucleic Acids Res. 2011 Dec;39(22):e153. doi: 10.1093/nar/gkr770. Epub 2011 Oct 5.
The identification of human cancer-related microRNAs (miRNAs) is important for cancer biology research. Although several identification methods have achieved remarkable success, they have overlooked the functional information associated with miRNAs. We present a computational framework that can be used to prioritize human cancer miRNAs by measuring the association between cancer and miRNAs based on the functional consistency score (FCS) of the miRNA target genes and the cancer-related genes. This approach proved successful in identifying the validated cancer miRNAs for 11 common human cancers with area under ROC curve (AUC) ranging from 71.15% to 96.36%. The FCS method had a significant advantage over miRNA differential expression analysis when identifying cancer-related miRNAs with a fine regulatory mechanism, such as miR-27a in colorectal cancer. Furthermore, a case study examining thyroid cancer showed that the FCS method can uncover novel cancer-related miRNAs such as miR-27a/b, which were showed significantly upregulated in thyroid cancer samples by qRT-PCR analysis. Our method can be used on a web-based server, CMP (cancer miRNA prioritization) and is freely accessible at http://bioinfo.hrbmu.edu.cn/CMP. This time- and cost-effective computational framework can be a valuable complement to experimental studies and can assist with future studies of miRNA involvement in the pathogenesis of cancers.
鉴定人类癌症相关 microRNAs(miRNAs)对于癌症生物学研究非常重要。尽管已经有几种鉴定方法取得了显著的成功,但它们忽略了与 miRNAs 相关的功能信息。我们提出了一种计算框架,可以通过基于 miRNA 靶基因和癌症相关基因的功能一致性评分(FCS)来测量癌症和 miRNAs 之间的关联,从而优先鉴定人类癌症 miRNAs。该方法在鉴定 11 种常见人类癌症的验证癌症 miRNAs 方面取得了成功,ROC 曲线下面积(AUC)范围从 71.15%到 96.36%。与 miRNA 差异表达分析相比,FCS 方法在鉴定具有精细调控机制的癌症相关 miRNAs 方面具有显著优势,例如结直肠癌中的 miR-27a。此外,一项甲状腺癌的案例研究表明,FCS 方法可以揭示新的癌症相关 miRNAs,例如 miR-27a/b,qRT-PCR 分析显示其在甲状腺癌样本中显著上调。我们的方法可以在基于网络的服务器 CMP(癌症 miRNA 优先级排序)上使用,并可在 http://bioinfo.hrbmu.edu.cn/CMP 上免费获得。这种节省时间和成本的计算框架可以作为实验研究的有益补充,并有助于未来研究 miRNA 在癌症发病机制中的作用。