Skip to main content

Table 7 Classification Performance with Reference Genomes of genus Equus

From: Machine learning on alignment features for parent-of-origin classification of simulated hybrid RNA-seq

Equus DNA

A

HiSat2

B

Hi_AS

C

Hi_RF

D

STAR

E

St_AS

F

St_RF

G

bwa

Accuracy

Sensitivity

Specificity

Precision

F1-score

MCC

AUPRC

AUROC

Pos Pref

Ties

79.1%

78.4%

79.9%

79.6%

79.0%

0.582

–

–

49.3%

–

78.9%

78.2%

79.4%

79.2%

78.7%

0.576

–

–

49.4%

33.8%

84.0%

90.6%

77.4%

80.0%

85.0%

0.686

93.7%

93.5%

56.6%

–

79.2%

78.6%

79.9%

79.6%

79.1%

0.584

–

–

49.4%

–

78.2%

77.5%

79.0%

78.6%

78.0%

0.564

–

–

49.3%

37.0%

86.0%

81.9%

90.1%

89.2%

85.4%

0.723

94.0%

93.9%

45.9%

–

88.9%

91.5%

86.3%

87.0%

89.2%

0.780

–

–

52.9%

–

  1. By many measures including accuracy, F1, and MCC, the random forest performance surpassed that of the other methods tested. For directional statistics, donkey and horse were considered the negative and positive classes, respectively. (A, B, C) Parent chosen by the HiSat2 aligner, or by comparing HiSat2 alignment scores, or by the random forest using HiSat2 alignment features, respectively. (D, E, F) Similar to columns A, B, and C but using the STAR aligner, configured for splicing. (G) Parent chosen by bwa