AUTHOR=Konganti Kranti , Reed Elizabeth , Mammel Mark , Kayikcioglu Tunc , Binet Rachel , Jarvis Karen , Ferreira Christina M. , Bell Rebecca L. , Zheng Jie , Windsor Amanda M. , Ottesen Andrea , Grim Christopher J. , Ramachandran Padmini 

TITLE=bettercallsal: better calling of Salmonella serotypes from enrichment cultures using shotgun metagenomic profiling and its application in an outbreak setting

JOURNAL=Frontiers in Microbiology

VOLUME=Volume 14 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/microbiology/articles/10.3389/fmicb.2023.1200983

DOI=10.3389/fmicb.2023.1200983

ISSN=1664-302X

ABSTRACT=Most Salmonella subtyping analyses now rely upon whole-genome sequencing (WGS) which focuses on the highresolution analysis of single genomes, or multiple single genomes, from isolated colonies on microbiological agar plates. Here we introduce bioinformatics innovations for a metagenomic outbreak response workflow that accurately identifies multiple Salmonella serovars at the same time. bettercallsal is one of the first analysis tools that can identify multiple Salmonella enterica serotypes from metagenomic or quasi-metagenomic data sets with high accuracy, allowing these isolate-independent methods to be incorporated into surveillance and root cause investigations. bettercallsal was tested against an in-silico benchmark data set, comprising 29 unique Salmonella serovar, 46 non-Salmonella bacterial, and 10 viral genomes at varying read depths, and on previously wellcharacterized and sequenced nonselective primary and selective enrichments of papaya and peach samples from separate outbreak investigations that resulted in identification of multiple Salmonella serovars using traditional isolate culturing and WGS as well as nucleic acid assays. Analyses were also performed on these data sets using a custom-built k-mer tool, SeqSero2, and Kallisto, to compare serotype calling to bettercallsal. The in-silico data set analyzed with bettercallsal achieved a maximum precision, recall and accuracy of 100%, 83% and 94% respectively.In the papaya outbreak samples, bettercallsal identified the presence of multiple serovars in concordance with Luminex xMAP assay results and also identified more serovars per sample as evidenced by the NCBI SNP clustering.In peach outbreak samples, bettercallsal identified two serovars in concordance with k-mer analysis and the Luminex xMAP assay. The genome hit reported by bettercallsal clustered with the isolate genome from chicken as reported by the FDA investigation of the peach outbreak from sequenced isolates (WGS). Overall, bettercallsal outperformed k-mer, Seqsero2, and Kallisto in identifying multiple serovars from enrichment cultures using shotgun metagenomics sequencing.