Many scientific works regard the possible use of miRNA levels as diagnostic and prognostic tools for different kinds of cancer. Developing reliable classifiers requires tackling several crucial aspects, some of which have been widely overlooked in the scientific literature: distribution of the measured miRNA expressions; statistical uncertainty that affects the parameters characterizing a classifier. Starting from a model problem, i.e. the classification of lung's adenocarcinomas and squamous cell carcinomas by relying on the expression of miR-205, miR-21 and snRNA U6, a new Bayesian classifier was developed, and its performance carefully analyzed. The method was first published in:
L. Ricci, V. Del Vescovo, C. Cantaloni, M. Grasso, M. Barbareschi and M. A. Denti Statistical analysis of a Bayesian classifier based on the expression of miRNAs, BMC Bioinformatics 16 (2015), 287, doi:10.1186/s12859-015-0715-9
The classifier was then used in several other context, for example in studies addressing frontotemporal dementia. Two representative publications are the following ones:
The classifier was implemented in a software package for the R environment, MiRNAQCD (miRNA quality control and diagnosis), available in at this GitHub repository, as well as on CRAN. The package is thoroughly described in:
M. Castelluzzo, A. Perinelli, S. Detassis, M. A. Denti and L. Ricci, MiRNA-QC-and-Diagnosis: An R package for diagnosis based on MiRNA expression, SoftwareX 12 (2020), 100569, doi:10.1016/j.softx.2020.100569