Datasets
Positive set: This dataset consists of miRNA sequences of 8 different localizations i.e., 16 axon, 69 circulating, 67 cytoplasm, 524 exosome, 25 extracellular vesicle, 21 microovesicle, 191 mitochondrion and 42 nucleus. The dataset can be downloaded here.
Negative dataset: This dataset comprises the miRNA sequences for the negative set corresponding to each localization. There are 830, 775, 808, 415, 829, 818, 659 and 799 sequences (nagetive set) for the localizations axon, circulating, cytoplasm, exosome, extracellular vesicle, microovesicle, mitochondrion and nucleus respectively. This daset can be retrieved here.
Negative dataset (miRBase): This negative dataset contains 951 miRNA sequences randomly taken from miRBase, where none of the two sequences share >80% degree of sequence similarity (ensured with CD-HIT program). This dataset is available here.
Independent test dataset: This dataset comprises 691 sequences with each sequence beloned to more than one localisations, unlike the positive dataset where each sequence belonged to one localisation only. This dataset is available here.