Skip to main content

Table 2 Average performance of models trained to classify into hominidae or one of the listed clades. The best 100 features were selected based on information gain and training/testing was performed with a 10/80/20 split at 100-fold MCCV

From: MicroRNA categorization using sequence motifs and k-mers

Hominidae
vs.
Holdout Test
F-measure Accuracy MCC F-measure Accuracy MCC
Hexapoda 0.93 0.93 0.86 0.93 0.93 0.86
Brassicaceae 0.82 0.93 0.78 0.92 0.92 0.84
Monocotyle 0.88 0.92 0.83 0.91 0.91 0.82
Nematoda 0.87 0.91 0.80 0.90 0.90 0.80
Fabaceae 0.81 0.88 0.72 0.87 0.87 0.75
Pisces 0.80 0.86 0.70 0.86 0.86 0.72
Virus 0.44 0.83 0.43 0.82 0.82 0.64
Aves 0.59 0.75 0.41 0.72 0.72 0.45
Laurasiatheria 0.54 0.73 0.39 0.70 0.72 0.45
Rodentia 0.62 0.69 0.37 0.69 0.69 0.38
Homo sapiens 0.62 0.61 0.23 0.62 0.61 0.23
Cercopithecidae 0.26 0.51 0.01 0.50 0.50 0.01
  1. Note, that for the test Hominidae vs H. sapiens the H. sapiens examples were removed from Hominidae. Table is sorted according to average model accuracy. This table presents average accuracy values, but Additional file 2: Figures S3-S5 present the accuracy distributions for 100 fold MCCV