Urinary microRNA ensembles for cancer detection and localization
MicroRNA ensembles formed by some microRNA species differ in healthy and cancer groups of people, and they represent a warning sign for various cancer scenarios. Since all microRNAs in blood cannot be transferred from donor to recipient cells during a single blood circulation, kidney filtration could pass some untransferred microRNAs from blood to urine. Here, we successfully identify uncommon microRNA ensembles in urine through a combination of nanowire-based microRNA extraction and machine learning-based analysis. By the nanowire-based extraction about 2500 species of urinary miRNAs were discovered, meaning that urine includes almost all human microRNA species. The machine learning-based analysis identified uncommon and suitable microRNA ensembles from the extracted microRNA species. The ensembles consisting of ~20 microRNA species could classify lung cancer, brain tumor, and non-cancer subjects with >95% accuracy, sensitivity, and specificity, even though they were early-staged.