Rapid geographical source attribution of Salmonella enterica serovar Enteritidis genomes using hierarchical machine learning
- University of Bristol, United Kingdom
- University of Bath, United Kingdom
- UK Health Security Agency, United Kingdom
- Utrecht University, Netherlands
Abstract
Salmonella enterica serovar Enteritidis is one of the most frequent causes of Salmonellosis globally and is commonly transmitted from animals to humans by the consumption of contaminated foodstuffs. In the UK and many other countries in the Global North, a significant proportion of cases are caused by consumption of imported food products or contracted during foreign travel, therefore making the rapid identification of the geographical source of new infections a requirement for robust public health outbreak investigations. Herein, we detail the development and application of a hierarchical machine learning model to rapidly identify and trace the geographical source of S. Enteritidis infections from whole genome sequencing data. 2,313 S. Enteritidis genomes, collected by the UKHSA between 2014-2019, were used to train a 'local classifier per node' hierarchical classifier to attribute isolates to 4 continents, 11 sub-regions and 38 countries (53 classes). The highest classification accuracy was achieved at the continental level followed by the sub-regional and country levels (macro F1: 0.954, 0.718, 0.661 respectively). A number of countries commonly visited by UK travellers were predicted with high accuracy (hF1: >0.9). Longitudinal analysis and validation with publicly accessible international samples indicated that predictions were robust to prospective external datasets. The hierarchical machine learning framework provided granular geographical source prediction directly from sequencing reads in <4 minutes per sample, facilitating rapid outbreak resolution and real-time genomic epidemiology. The results suggest additional application to a broader range of pathogens and other geographically structured problems, such as antimicrobial resistance prediction, is warranted.
Data availability
The final optimised hierarchical model as well as a pipeline for pre-processing raw read data to unitigs/patterns for input and paper data is available from https://github.com/SionBayliss/HierarchicalML with a short description and tutorial for ease of use. This end-to-end process, from FASTQ to prediction, is open access and available to users under GNU GPL licence . This depository also includes the preprocessed unitig datasets and resulting predictions. Short read sequencing data is available from the Sequence Read Archive (Bioproject: PRJNA248792). Please not that the sequence data has been previously deposited/published in the Sequence Read Archive by PHE/UKHSA and was not generated for this project.
-
Public Health England - SalmonellaNCBI SRA: PRJNA248792.
Article and author information
Author details
Funding
Academy of Medical Sciences (SBF005\1089)
- Lauren A Cowley
The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.
Reviewing Editor
- Ben S Cooper, University of Oxford, United Kingdom
Version history
- Preprint posted: August 25, 2022 (view preprint)
- Received: October 13, 2022
- Accepted: April 2, 2023
- Accepted Manuscript published: April 12, 2023 (version 1)
- Version of Record published: April 28, 2023 (version 2)
Copyright
© 2023, Bayliss et al.
This article is distributed under the terms of the Creative Commons Attribution License permitting unrestricted use and redistribution provided that the original author and source are credited.
Metrics
-
- 1,109
- views
-
- 147
- downloads
-
- 5
- citations
Views, downloads and citations are aggregated across all versions of this paper published by eLife.
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Rapid geographical source attribution of Salmonella enterica serovar Enteritidis genomes using hierarchical machine learning
eLife 12:e84167.
https://doi.org/10.7554/eLife.84167
Further reading
-
- Cancer Biology
- Epidemiology and Global Health
Background:
Age is the most important risk factor for cancer, but aging rates are heterogeneous across individuals. We explored a new measure of aging-Phenotypic Age (PhenoAge)-in the risk prediction of site-specific and overall cancer.
Methods:
Using Cox regression models, we examined the association of Phenotypic Age Acceleration (PhenoAgeAccel) with cancer incidence by genetic risk group among 374,463 participants from the UK Biobank. We generated PhenoAge using chronological age and nine biomarkers, PhenoAgeAccel after subtracting the effect of chronological age by regression residual, and an incidence-weighted overall cancer polygenic risk score (CPRS) based on 20 cancer site-specific polygenic risk scores (PRSs).
Results:
Compared with biologically younger participants, those older had a significantly higher risk of overall cancer, with hazard ratios (HRs) of 1.22 (95% confidence interval, 1.18–1.27) in men, and 1.26 (1.22–1.31) in women, respectively. A joint effect of genetic risk and PhenoAgeAccel was observed on overall cancer risk, with HRs of 2.29 (2.10–2.51) for men and 1.94 (1.78–2.11) for women with high genetic risk and older PhenoAge compared with those with low genetic risk and younger PhenoAge. PhenoAgeAccel was negatively associated with the number of healthy lifestyle factors (Beta = –1.01 in men, p<0.001; Beta = –0.98 in women, p<0.001).
Conclusions:
Within and across genetic risk groups, older PhenoAge was consistently related to an increased risk of incident cancer with adjustment for chronological age and the aging process could be retarded by adherence to a healthy lifestyle.
Funding:
This work was supported by the National Natural Science Foundation of China (82230110, 82125033, 82388102 to GJ; 82273714 to MZ); and the Excellent Youth Foundation of Jiangsu Province (BK20220100 to MZ).
-
- Epidemiology and Global Health
Background:
Circulating omega-3 and omega-6 polyunsaturated fatty acids (PUFAs) have been associated with various chronic diseases and mortality, but results are conflicting. Few studies examined the role of omega-6/omega-3 ratio in mortality.
Methods:
We investigated plasma omega-3 and omega-6 PUFAs and their ratio in relation to all-cause and cause-specific mortality in a large prospective cohort, the UK Biobank. Of 85,425 participants who had complete information on circulating PUFAs, 6461 died during follow-up, including 2794 from cancer and 1668 from cardiovascular disease (CVD). Associations were estimated by multivariable Cox proportional hazards regression with adjustment for relevant risk factors.
Results:
Risk for all three mortality outcomes increased as the ratio of omega-6/omega-3 PUFAs increased (all Ptrend <0.05). Comparing the highest to the lowest quintiles, individuals had 26% (95% CI, 15–38%) higher total mortality, 14% (95% CI, 0–31%) higher cancer mortality, and 31% (95% CI, 10–55%) higher CVD mortality. Moreover, omega-3 and omega-6 PUFAs in plasma were all inversely associated with all-cause, cancer, and CVD mortality, with omega-3 showing stronger effects.
Conclusions:
Using a population-based cohort in UK Biobank, our study revealed a strong association between the ratio of circulating omega-6/omega-3 PUFAs and the risk of all-cause, cancer, and CVD mortality.
Funding:
Research reported in this publication was supported by the National Institute of General Medical Sciences of the National Institute of Health under the award number R35GM143060 (KY). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
