[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-123623-en":3,"doc-seo-123623-105":30,"detail-sidebar-cat-0-en-105":83},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},123623,687197100911,"Himbo","https://ap-avatar.wpscdn.com/avatar/a000239b6f1da00475?x-image-process=image/resize,m_fixed,w_180,h_180&k=1785132997149421697",8,"Research & Report","Rapid geographical source attribution of Salmonella enterica serovar Enteritidis genomes using hierarchical machine learning","Rapid geographical source attribution of Salmonella enterica serovar Enteritidis genomes is addressed using a hierarchical machine learning (hML) framework. The approach models a geographic hierarchy aligned with the UN M49 regional codes and evaluates multiple co-optimized combinations of model type, resampling strategy, and feature selection. Performance is summarized using hierarchical F1 measures and classification metrics on a held-out test set, while genomic diversity across countries is quantified using SNP-based linkage clustering. Results support region- and country-level risk and sampling-effort comparisons for 2014–2019 isolates.","RESEARCH ARTICLE  \nFigures and figure supplements  \nRapid geographical source attribution of Salmonella enterica serovar Enteritidis genomes using hierarchical machine learning  \nSion C Bayliss et al.  \n Research article Epidemiology and Global Health | Microbiology and Infectious Disease  \nFigure 1. Summary of S. Enteritidis isolates collected by the UKHSA from UK clinical patients who recently reported foreign travel between 2014–2019.(A) Geographical distribution of 2,313 S. Enteritidis isolates by reported foreign travel. Variably sized points represent the number of samples per country. The map is colored by region (Africa: yellow, Americas: red, Asia: purple, Europe: blue) . (B) Maximum likelihood phylogenetic tree of 2,313 S. Enteritidis isolates with bar colored by region of origin. (C) Kernel density plot indicating sampling density per region through time. No correction was Figure 1 continued on next page  \nBayliss et al. eLife 2023;12:e84167. DOI: [https://doi.org/10.7554/eLife.84167](https://doi.org/10.7554/eLife.84167) 2 of 15  \n Research article Epidemiology and Global Health | Microbiology and Infectious Disease  \nFigure 1 continued  \nmade for the seasonal variation observed in international travel. (D) Comparison of the consistency of the sampling effort of the UKHSA to all publicly available S. Enteritidis isolates on NCBI for the same period. Isolates were resampled to control for variable sample number per year and compared toa uniform distribution using the Kolmogorov-Smirnov D statistic (NCBI = red, UKHSA = blue) . Higher values indicate greater deviation from a uniform distribution. (E) Relative risk per country of acquiring S. Enteritidis infection when traveling. A risk score was generated by dividing the proportion of UKHSA clinical isolates per country by the proportion of all UK travelers traveling to that country as recorded by the Office of National Statistics (ONS) . Only countries present in both datasets were used to calculate proportions.  \nBayliss et al. eLife 2023;12:e84167. DOI: [https://doi.org/10.7554/eLife.84167](https://doi.org/10.7554/eLife.84167) 3 of 15  \n Research article Epidemiology and Global Health | Microbiology and Infectious Disease  \nFigure 1—figure supplement 1. Summary of 2,313 S. Enteritidis isolates collected by the UKHSA from UK clinical patients during 2014-2019 who recently reported foreign travel to 38 individual country classes. Each panel contains a bar chart of isolate counts per year per country class.  \nBayliss et al. eLife 2023;12:e84167. DOI: [https://doi.org/10.7554/eLife.84167](https://doi.org/10.7554/eLife.84167) 4 of 15  \n Research article Epidemiology and Global Health | Microbiology and Infectious Disease  \nFigure 2. Summary statistics showing model and resampling scheme selection, feature selection, and optimization of the S. Enteritidis hML source attribution model. (A) Example schematic of a geographical hierarchy based upon the UN M49 Standard for regional codes. (B) Table of summary statistics for the ten top-performing co-optimized model and resampling methods from a cohort of 36 combinations, sorted by hF1 (high-low) . Training time is reported in the final column in seconds. A black box indicates the top four models used for feature selection. (C) Grouped bar chart comparing Figure 2 continued on next page  \nBayliss et al. eLife 2023;12:e84167. DOI: [https://doi.org/10.7554/eLife.84167](https://doi.org/10.7554/eLife.84167) 5 of 15  \n Research article Epidemiology and Global Health | Microbiology and Infectious Disease  \nFigure 2 continued  \nmacro F1 per hierarchical level for the ten top-performing model/resampled combinations. (D) Table of summary statistics for random forest featureselection applied to the four top-performing co-optimized model and resampling methods. Black boxes indicate the optimal number of features per combination. (E) Grouped bar chart comparing macro F1 per hierarchical level for the four top-performing co-optimied model a","cbCaifVI7v4lCPaT","https://ap.wps.com/l/cbCaifVI7v4lCPaT","pdf",9228600,1,15,"English","en",105,"# Figures and figure supplements\n## Summary of S. Enteritidis isolates and travel-based geography (Figure 1)\n## Hierarchical machine learning model design and selection (Figure 2)\n## Classification results and genomic diversity per country (Figure 3)\n## Additional summary for Figure 1 (Figure 1—figure supplement 1)","[{\"question\":\"How are model performance and diversity assessed?\",\"answer\":\"Performance is evaluated with hierarchical F1 (hF1) and related classification metrics on a test dataset, while genomic diversity is estimated using SNP-based single-linkage cluster counts normalized by sample number per country.\"}]","Rapid geographical source attribution of Salmonella enterica serovar Enteritidis genomes using hierarchical machine learning | PDF",1785817692,38,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":78,"head_meta":80,"extra_data":82,"updated_unix":28},"rapid-geographical-source-attribution-of-salmonella-enterica-serovar-enteritidis-genomes-using-hierarchical-machine-learning","",{"@graph":36,"@context":77},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/rapid-geographical-source-attribution-of-salmonella-enterica-serovar-enteritidis-genomes-using-hierarchical-machine-learning/123623/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71],{"name":72,"@type":73,"acceptedAnswer":74},"How are model performance and diversity assessed?","Question",{"text":75,"@type":76},"Performance is evaluated with hierarchical F1 (hF1) and related classification metrics on a test dataset, while genomic diversity is estimated using SNP-based single-linkage cluster counts normalized by sample number per country.","Answer","https://schema.org",{"og:url":52,"og:type":79,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":81,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":84},[85,89,93,97,102,107,112,115,120,123,127],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":86,"show_sort_weight":87,"slug":88},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":90,"show_sort_weight":91,"slug":92},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Exam",70,"exam",{"id":98,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},5,"Comic",60,"comic",{"id":103,"doc_module":4,"doc_module_name":46,"category_name":104,"show_sort_weight":105,"slug":106},6,"Technology",50,"technology",{"id":108,"doc_module":4,"doc_module_name":46,"category_name":109,"show_sort_weight":110,"slug":111},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":113,"slug":114},30,"research-report",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},9,"Religion & Spirituality",20,"religion-spirituality",{"id":118,"doc_module":4,"doc_module_name":46,"category_name":121,"show_sort_weight":118,"slug":122},"World Cup","world-cup",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":124,"slug":126},10,"Lifestyle","lifestyle",{"id":128,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":98,"slug":130},19,"General","general"]