A Data Mining Strategy Identifies microRNA-15b-5p as a Potencial Bio-Marker in Non-Ischemic Heart Failure

Advances in bioinformatics has provided us with large databases that characterize the complex genetic and epigenetic changes associated with human diseases. The use of data mining strategies on public access databases to identify previously unknown disease markers is an innovative approach to identify potential biomarkers or even new therapeutic targets in complex diseases such as heart failure (HF). We assessed the hypothesis that a bioinformatic strategy of meta-analysis using existing public databases may predict novel differentially expressed microRNA (miR) in non-ischemic HF patients.We systematically reviewed the literature for miR profiling and HF (2006-2014). Four studies fulfilled quality criteria for analysis. Raw data from the largest study (discovery dataset, n=35) was normalized and analyzed by unsupervised hierarchical clustering. Significance analysis selected miRs with a fold change of 2 and a false discovery rate of less that 1%. The miRs reported as significant in the other three papers were incorporated in a meta-analysis using a robust rank aggregator approach and then contrasted with the results of the unsupervised analysis of raw data. To validate the in-silico analysis, we evaluated the miRs with higher predicted fold change in plasma samples from 23 patients with HF and five healthy controls using RT-qPCR. Internal controls included miR-39 for standardization of the extraction procedures and miR23a, miR451 as hemolysis markers. Results: The in-silico strategy identified 54 miR that fulfilled our criteria. The meta-analysis confirmed nine differentially expressed miRs, six of them previously reported (let-7b, miR-100, miR-103, miR-199a, miR-23a) and three with no known relation with HF (miR-125b, miR-140, miR-15b). RT-qPCR of HF plasma samples revealed that miR-15b-5p was significantly reduced in HF subjects (p=0.004). Conversely, the previously described miR-23a (p = 0.0021) and let-7b (p = 0.0195) were largely increased in HF subjects. In conclusion, bioinformatics analysis allows the identification of previously unreported miR associated with HF. This novel approach using publicly available data for the identification of new potential biomarkers may accelerated the pre-analytic phase of biomarker research.