Fig. 1
TyphiNET data curation and dashboard architecture. A The Global Typhoid Genomics Consortium (GTGC) aggregates and curates Typhi genome data and metadata, using Pathogenwatch as both an analysis platform (calling genotypes and AMR determinants from genome assemblies) and publicly accessible data store. Metadata that are not typically available in NCBI/ENA but are collected and curated by the GTGC include purpose of sampling (to tag datasets that are suitable for estimating AMR/genotype prevalence) and information on country-of-travel for travel-associated isolates (to identify country of origin). B A web-scraper is used to pull the latest versions of genotypes, AMR determinants, and metadata files from GTGC-curated Typhi collections in Pathogenwatch, which are used to populate the TyphiNET database. C The TyphiNET dashboard is implemented as a MERN (MongoDB, Express, React, Node) stack JavaScript application as illustrated. Genome data are filtered to exclude low-quality genome sequences, and data sets whose sampling frames make them unsuitable for AMR surveillance (such as those targeted towards sequencing of resistant strains only, or outbreak investigations), before calculating national/annual prevalences of AMR and genotypes to display in interactive plots. ReactJS is used to provide user interface layouts suitable for viewing the interactive plots on a range of devices (computer, tablet, phone). Users can also download static images of current plot displays (PNG), static reports with all current plots (PDF format), or a copy of the TyphiNET database (CSV format)