Table 1 Average performance of the different methods across multiple metrics.

From: Gene expression analysis method integration and co-expression module detection applied to rare glucide metabolism disorders using ExpHunterSuite

 

FP

FN

Precision

Recall

F-measure

AUC

Combined

4.33

3.33

0.86

0.89

0.88

0.98

DESeq2

4.33

3.00

0.87

0.90

0.88

0.98

edgeR

4.33

3.67

0.86

0.88

0.87

0.96

limma

2.33

10.67

0.90

0.66

0.76

0.98

NOISeq

17.33

3.33

0.65

0.89

0.74

0.96

Threshold 1

17.33

3.00

0.65

0.90

0.75

–

Threshold 2

4.33

3.00

0.87

0.90

0.88

–

Threshold 3

4.33

3.67

0.86

0.88

0.87

–

Threshold 4

4.33

3.67

0.86

0.88

0.87

–

Naïve Bayes

4.33

3.00

0.87

0.90

0.88

–

  1. FP false positives, FN false negatives, AUC area under the curve, recall sensitivity, precision positive predictive value (PPV). AUC is not shown for the threshold and Naïve Bayes methods as only single values of sensitivity and specificity could be produced with these methods. Numbers in bold represente the highest values for each metric.
Back to article page