miMETA: an online meta-analysis tool for the miRandola database

MicroRNAs (miRNAs) are small non-coding RNAs that act as post-trancriptional regulators of protein coding genes. A lot of miRNAs have been found to correlate well with many human cancers and other diseases. Recently, miRNAs have been found in human body fluids such as plasma, serum and urine and they are considered potential non-invasive biomarkers. For this reason we recently developed miRandola, the first extracellular circulating miRNAs database. To improve the usability of the database as a key source to inform clinical practice, in the present study we introduce miMETA, a meta-analysis tool for miRandola.miMETA incorporates two R packages, Mada ('Meta-Analysis of Diagnostic Accuracy') and Metafor ('Meta-Analysis Package for R'). Mada provides functions for diagnostic meta-analysis, Metafor provides a comprehensive collection of functions for conducting meta-analysis in R. Standard methods for diagnostic accuracy meta-analysis have been applied: sensitivity, specificity, diagnostic odds ratio (DOR), log odds ratio, and the area under the curve (AUC). The AUC represents an analytical performance summary displaying sensitivity-specificity trade off. The methodological quality of each study can be assessed by QUADAS (QUality Assessment for studies of Diagnostic Accuracy), a 14-questions evidence-based quality assessment tool used for diagnostic accuracy studies evaluation. When a criterion is fulfilled, unclear or not achieved a score of 1, 0 or -1 is assigned respectively. Only studies with higher QUADAS score (>= 10) are accepted. Users may select a disease related to extracellular miRNAs annotated in the miRandola database. Next they may decide to apply the QUADAS test. Finally information on the number of patients with True Positive (TP), False Positive (FP), True Negative (TN), False Negative (FN) tests and related miRNAs, for each study may be uploaded.Availability: http://atlas.dmi.unict.it/mirandola/mimeta.php