Development and validation of the Mining_RNA system

Abstract

The human genome sequencing provided the deepening of several types of studies and technologies of biological analysis, among these, the microarray. The need to publish research raw data boosted the creation of public databases where this information could be indexed and retrieved. These databases are a great source of transcriptomic data that unfortunately end up being underutilized. Our aim was the development of a WEB system for data mining in transcriptomic studies from microarrays stored in the Gene Expression Omnibus biological database (GEO). Through a step-by-step usability together with a series of filters the Mining_RNA system makes it possible to retrieve GEO data, calculate the differential expression between the genes of a study, and statistical analysis for each gene of the analyzed study. The system was validated by comparing with the evaluation of the same data with the GEO2R software (approximately 98% effectiveness) and in the original study (effectiveness higher than 90%). Mining_RNA can be a strong ally for researchers for the re-analysis of transcriptomic studies, enabling a new way of analyzing data and generating results as reliable as already consolidated tools.