Skip to main content

Table 4 Classification Performance with Reference Transcripts of genus Brassica

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

Brassica RNA

A

Bowtie2

B

Bo_AS

C

Bo_RF

D

STAR

E

St_AS

F

St_RF

G

Salmon

H

bwa

Accuracy

89.3%

92.1%

93.9%

91.8%

92.2%

93.8%

89.4%

89.6%

Sensitivity

92.8%

94.7%

92.8%

94.3%

94.6%

93.7%

89.3%

92.9%

Specificity

85.9%

89.5%

94.9%

89.2%

89.8%

93.9%

89.6%

86.3%

Precision

86.8%

90.1%

94.8%

89.7%

90.3%

93.9%

89.6%

87.1%

F1-score

89.7%

92.3%

93.8%

92.0%

92.4%

93.8%

89.4%

89.9%

MCC

0.789

0.844

0.877

0.836

0.846

0.876

0.788

0.793

AUPRC

–

–

–

–

–

98.4%

–

–

AUROC

–

–

98.4%

–

–

98.4%

–

–

Pos Pref

53.4%

52.6%

49.0%

52.6%

52.4%

49.9%

49.8%

53.3%

Ties

–

3.3%

–

–

5.0%

–

–

–

  1. Performance metrics for transcript-based parent-of-origin classification in Brassica. Whether used with Bowtie2 or STAR, the random forest improved the accuracy, F1, and MCC. For directional statistics, species B. rapa and B. oleracea were considered the negative and positive classes, respectively. (A, B, C) Using the Bowtie2 aligner, a parent was chosen by the aligner, or by comparing alignment scores, or by the random forest, respectively. (D, E, F) Similar to A, B, and C but using the STAR aligner, configured to avoid splicing. (G, H) Parent chosen by Salmon or bwa, respectively