Skip to main content

Table 4 Datasets used in this study. For the number of strains and SNPs, the final numbers after all filtering are provided

From: Machine learning and phylogenetic analysis allow for predicting antibiotic resistance in M. tuberculosis

Drug name

Line of therapy

Pharmacological group

Number of strains

Number (fraction) of resistant strains

Number of SNPs (features)

STR

First line

Aminoglycosides

4,726

1158 (24,5%)

24,425

AMK

Second line

Aminoglycosides

1,149

208 (18,1%)

18,864

CAP

Second line

Aminoglycosides

1,086

205 (18,9%)

17,045

KAN

Second line

Aminoglycosides

1,362

297 (21,8%)

17,335

OFL

Second line

Fluoroquinolones

795

307 (38,6%)

14,185

ETH

Second line

Nicotinamide derivative

571

210 (36,8%)

12,974

Total

4,869

 

24,425