TABLE 3.

Comparison of nine examined genotyping methods in this study^a

Assessment indicator	non-WGS-based typing			WGS-based typing
	Low-resolutionb		High-resolution	Medium-resolution				High-resolution
	7-gene legacy MLST	CRISPR	CRISPR-MVLST	cgMLST	Plasmid profile	AMR gene profile	VF gene profile	CoreSNP	wgMLST
Repeatability	Excellent	Excellent	Excellent	Excellent	Excellent	Excellent	Excellent	Excellent	Excellent
Reproducibility	Excellent	Excellent	Excellentc	Excellent	Excellent	Excellent	Excellent	Excellent	Excellent
Discriminatory power (DI)d	0.8158	0.8377	0.9628	0.8541	0.8852	0.9361	0.9497	0.9967	1
Discriminatory types	11	15	30	18	18	32	36	56	61
Scheme standardized or note	Yes	Yes	Yes	Yes	No	No	No	Not yet	Not yet
Ease of interpretation of data generatedf	Excellent	Excellent	Excellent	Good	Excellent	Excellent	Excellent	Moderate	Moderate
Ease of use	Good to moderateg	Moderate	Moderate	Moderate	Good	Good	Good	Poor	Moderate
High throughput	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Cost	Moderate	Low	Low	Moderate	Moderate	Moderate	Moderate	Moderate	Moderate
Time required (days)h	1~2	1~2	1~2	2~7	2~7	2~7	2~7	2~7	2~7

^aRankings 1. DI in order (good to poor): MIC profile = wgMLST > CoreSNP > AMR profile > CRISPR-MVLST > VF gene profile > AMR gene profile > plasmid profile > cgMLST > CRISPR > MLST > serotyping. 2. Easy of use (good to poor): CRISPR > CRISPR-MVLST > MLST > cgMLST = wgMLST > plasmid profile = AMR gene profile = VF gene profile = CoreSNP > AMR profile = MIC profile = serotyping. 3. Cost (low to high): CRISPR < CRISPR-MVLST < MLST = VF gene profile = AMR gene profile = plasmid profile < wgMLST = cgMLST = coreSNP < AMR profile < serotyping < MIC profile. 4. Time required (short to long): MLST < CRISPR < CRISPR-MVLST < AMR profile = MIC profile < plasmid profile = VF gene profile = AMR gene profile < wgMLST < CoreSNP = cgMLST < serotyping.

^bFor reading ease, Low/Medium/High resolution classification is set up for genotyping methods according to their DI results in this study.

^cWe summarized nine methods in a similar manner of a previous study (28).

^dUsing the Discriminatory index (DI) for a description of discriminatory power.

^eIf there is 1~2 universally acknowledged standard for this typing or not.

^fIntended as unequivocal interpretation.

^gUsing in silico prediction will be easier.

^hThe approximate number of days to get typing results is estimated by excluding the interval of time to obtain a single pure colony suitable to be handled by the method.