Abstract 3995: Identification of putative miRNA targets by a systems biological approach in NCI-60 data

MicroRNAs (miRNAs) are single-stranded, non-coding RNA molecules and on average only 22 nucleotides long that regulate gene expression at the post-transcriptional level. Target predictions of miRNA based on sequence are limited due to too many candidates resulting in high false positives. To overcome this limitation functional target prediction approaches have been investigated by using mRNA and miRNA expression data. These approaches used in the computational assessment of functional miRNA-target association. However, these approaches with miRNA-mRNA association were not enough to verify real functional targets at protein level, which could affect on biological signaling pathway. Therefore, in order to overcome current limitation we developed a novel correlation-based approach (sequence-independence) that predict the association of miRNAs with signaling pathways by using genomics and proteomics data from NCI-60 cancer cell lines (mRNA, miRNA, and protein expression data). To identify miRNA-protein association, we used two correlation-based methods: direct comparison (miRNA-protein pairs) and indirect comparison (miRNA-mRNA and protein-mRNA pairs), then compared their outputs. In direct comparison, we calculated Pearson correlation coefficients for 96k probe pairs and found over 3 k significant correlated pairs (3.2 %) at the cut-off p-value < 0.005: among them 53.4 % were negative and 46.6% were positive. In indirect comparison, we ranked the number of correlated mRNA probes between miRNA-mRNA and protein-mRNA pairs, in which we calculated over 11 and 5.7 million probe pairs respectively. Then, two significant output pair sets were combined by intersection of mRNA probes and then determined the number of common probes. With these determined numbers, finally we generated association score matrix between miRNA and protein probes. Using those score matrix, we evaluated the highly associated miRNAs with 40 biological pathways, which play critical roles in cell cycle, proliferation and metastasis. In conclusion, we developed a noble expression profile-based approach to systematically predict potential miRNA functional targets which play critical roles in cell signaling pathways from the NCI-60 genomic and proteomic data at whole genomic scale. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 3995. doi:10.1158/1538-7445.AM2011-3995