MGS2AMR: a gene-centric mining of metagenomic sequencing data for pathogens and their antimicrobial resistance profile

Table 2 Instances used for building AMR XGBoost prediction models per antibiotic

Antibiotic	Class	Samples	ARG	S	R
Ampicillin	Beta-lactam	1376	28	57	1319
Cefepime	Beta-lactam	720	29	240	480
Gentamicin	Aminoglycoside	1550	52	560	990
Meropenem	Beta-lactam	1072	31	550	522
Tetracycline	Tetracycline	1333	7	446	887
Tobramycin	Aminoglycoside	1258	49	482	776
Trimethoprim-sulfamethoxazole	Sulfonamide	1356	3	329	1027
Vancomycin	Glycopeptide	95	7	15	80

Class represents the class of the antibiotic. Samples is the total number of samples available for training and testing the XGBoost models. ARG is the number of unique ARG found across these samples (and used as input). S and R are the numbers of samples susceptible or resistant to the AB, respectively, as defined by the clinical laboratory metadata (antibiogram)

ISSN: 2049-2618