The Escherichia coli species contains a diverse set of sequence types and there remain important questions regarding differences in genetic content within this population that need to be addressed. Pangenomes are useful vehicles for studying gene content within sequence types. Here, we analyse 21 E. coli sequence type pangenomes using comparative pangenomics to identify variance in both pangenome structure and content. We present functional breakdowns of sequence type core genomes and identify sequence types that are enriched in metabolism, transcription and cell membrane biogenesis genes. We also uncover metabolism genes that have variable core classification, depending on which allele is present. Our comparative pangenomics approach allows for detailed exploration of sequence type pangenomes within the context of the species. We show that ongoing gene gain and loss in the E. coli pangenome is sequence type-specific, which may be a consequence of distinct sequence type-specific evolutionary drivers.

The Escherichia coli species contains a diverse set of sequence types and there remain important questions regarding differences in genetic content within this population that need to be addressed. Pangenomes are useful vehicles for studying gene content within sequence types. Here, we analyse 21 E. coli sequence type pangenomes using comparative pangenomics to identify variance in both pangenome structure and content. We present functional breakdowns of sequence type core genomes and identify sequence types that are enriched in metabolism, transcription and cell membrane biogenesis genes. We also uncover metabolism genes that have variable core classification depending on which allele is present. Our comparative pangenomics approach allows for detailed exploration of sequence type pangenomes within the context of the species. We show that pangenome evolution is independent of phylogenetic signal at the phylogroup level, which may be a consequence of distinct sequence type-specific driving factors relating to ecology and pathogenic phenotype.Data SummarySupporting data and code have been provided within the article or through Supplementary Data files available at https://doi.org/10.6084/m9.figshare.19793758. Custom Python scripts used to perform analyses are available at github.com/lillycummins/InterPangenome unless otherwise stated in the text.