FUNCTIONALITY Lack of active curation/updates for ARG-ANNOT mean cannot be used for CP surveillance

[fmaguire]

Need a justification for why ARG-ANNOT was used.
As the author's point out genomic surveillance of CP genes is important, therefore using a database that has not been updated in two years is a problem as it won't capture the latest discovered CP diversity. For comparison, a crude grep -c of different beta-lactamase AMR gene names (mostly CP) shows for most that ARG-ANNOT lacks a lot of CP alleles retained in databases are updated multiple times per year.

db	version	VIM	OXA	NDM	KPC	IMP	IMI
CARD	3.0.9	39	439	27	19	47	6
ResFinder	2020-06-02	53	526	24	39	68	4
NCBI Bacterial AMR Reference Gene Database	2020-06-11.1	68	874	29	51	80	18
ARG-ANNOT/CPGeneProfiler	v4 May 2018	42	33	14	20	39	6

Highly, highly recommend modifying this tool to use NCBI, CARD, or ResFinder so that its CP repertoire is actively curated and regularly updated.

[ramadatta]

We completely agree with reviewer’s comment. Based on the reviewer's recommendation, we have replaced ARG-ANNOT CP gene database with a CP gene database extracted from a very recent version of NCBI Bacterial Antimicrobial Resistance Reference Gene Database (version 2020-07-16.2). The criteria on how the CP genes were selected from the NCBI database and the exclusion of non-carbapenemase OXA gene variants are also now mentioned in the manuscript.