TB-DROP: deep learning-based drug resistance prediction of Mycobacterium tuberculosis utilizing whole genome mutations

Table 4 The difference of metrics between the original models and the reimplementation version

Model	Metrics	Original	Reimplementation	Difference (%)
WDNN	Sum of AUC	10.50	10.50	0
DeepAMR	Sum of AUC	3.75	3.87	3.20
CNNGWP	MSE	62.34	63.04	1.12

The column ‘difference’ is calculated as (value of reimplementation – value of original) / value of original. AUC stands for “Area under the ROC Curve”; MSE stands for Mean Squared Error

ISSN: 1471-2164