Improved Pre-miRNA Classification by Reducing the Effect of Class Imbalance

Table 3

Datasets and detailed classification results of MiRNAClassify and HuntMi.

Method	Species	Dataset	Type	Size	SE (%)	SP (%)	(%)
MiRNAClassify	Homo sapiens	HM_hsa_positive_set	Real	1406	97.15	98.35	97.75
HuntMi		HM_hsa_negative_set	Pseudo	81228	95.02	96.94	95.98
MiRNAClassify	Animal	HM_animal_positive_set	Real	7053	95.74	97.55	96.64
HuntMi		HM_animal_negative_set	Pseudo	218154	94.11	95.95	95.03
MiRNAClassify	Plant	HM_plant_positive_set	Real	2172	93.32	97.82	95.54
HuntMi		HM_plant_negative_set	Pseudo	114929	91.71	95.87	93.77
MiRNAClassify	Homo sapiens	HM_hsa_updated_set	Real	445	93.93
HuntMi					92.14
MiRNAClassify	Arabidopsis thaliana	HM_ath_updated_set	Real	68	92.65
HuntMi					91.18
MiRNAClassify	Oryza sativa	HM_osa_updated_set	Real	169	75.74
HuntMi					69.82
MiRNAClassify	Glycine max	HM_gma_updated_set	Real	302	92.72
HuntMi					88.41