Background: The detection and identification of pathogenic factors (PFs) scattered across a genome can provide useful insights about the pathogenic potential of bacterial strains. Although very important, this process is often carried out in little depth, also due to a lack of tools that predict/identify potential pathogenic factors in poorly characterized species. We developed a collection of computational tools named ScrInHeat that allows to screen for PFs providing standardized annotation and visualization. Methods: ScrInHeat works through the following steps: building of PFs databases, screening of genomic assemblies/sets of proteins, integration of results, and generation of Gene Ontology (GO) annotated heatmaps for a simple results visualization. Its performance was evaluated through comparison against 3 freely available databases/tools with similar purpose: VFDB, PATRIC and ABRicate. Performances were compared using NCBI genomic assemblies of a well- and a poorly-characterized species, respectively Escherichia coli and Achromobacter xylosoxidans. Results: ScrInHeat was able to build reliable PFs databases when compared to the well-curated ones: although some PFs were not identified (62% content overlap), all the sequences within ScrInHeat database were truly related to pathogenicity. Moreover, when compared to the currently available A. xylosoxidans database, ScrInHeat could identify 148 additional PFs. ScrInHeat also offered cellular component, molecular function and biological process GO annotations while reducing the manual effort needed to filter and display results in the form of a heatmap. Conclusions: ScrInHeat proved to be a fast and versatile tool to identify PFs and readily visualize and interpret results with the support of GO annotations. Overall, it represents a new useful starting tool to study poorly characterized bacteria and aid microbiologists by prompting further characterization of putative PFs.
ScrInHeat: a computational approach for genomic screening of pathogenic factors in poorly studied bacteria / Veschetti, Laura; MORON DALLA TOR, Lucas; Lleò, Maria M.; Sandri, Angela; Malerba, Giovanni. - (2021), pp. 1-1. (Intervento presentato al convegno EMBO | EMBL Symposium: New Approaches and Concepts in Microbiology tenutosi a virtuale nel 07-09 Luglio 2021).
ScrInHeat: a computational approach for genomic screening of pathogenic factors in poorly studied bacteria
Lucas Moron Dalla Tor;
2021-01-01
Abstract
Background: The detection and identification of pathogenic factors (PFs) scattered across a genome can provide useful insights about the pathogenic potential of bacterial strains. Although very important, this process is often carried out in little depth, also due to a lack of tools that predict/identify potential pathogenic factors in poorly characterized species. We developed a collection of computational tools named ScrInHeat that allows to screen for PFs providing standardized annotation and visualization. Methods: ScrInHeat works through the following steps: building of PFs databases, screening of genomic assemblies/sets of proteins, integration of results, and generation of Gene Ontology (GO) annotated heatmaps for a simple results visualization. Its performance was evaluated through comparison against 3 freely available databases/tools with similar purpose: VFDB, PATRIC and ABRicate. Performances were compared using NCBI genomic assemblies of a well- and a poorly-characterized species, respectively Escherichia coli and Achromobacter xylosoxidans. Results: ScrInHeat was able to build reliable PFs databases when compared to the well-curated ones: although some PFs were not identified (62% content overlap), all the sequences within ScrInHeat database were truly related to pathogenicity. Moreover, when compared to the currently available A. xylosoxidans database, ScrInHeat could identify 148 additional PFs. ScrInHeat also offered cellular component, molecular function and biological process GO annotations while reducing the manual effort needed to filter and display results in the form of a heatmap. Conclusions: ScrInHeat proved to be a fast and versatile tool to identify PFs and readily visualize and interpret results with the support of GO annotations. Overall, it represents a new useful starting tool to study poorly characterized bacteria and aid microbiologists by prompting further characterization of putative PFs.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
