Support vector machine applied to predict the zoonotic potential of E. coli O157 cattle isolates

Nadejda Lupolova; Tim Dallman; Louise Matthews; James L. Bono; David L. Gally

doi:10.1073/pnas.1606567113

Support vector machine applied to predict the zoonotic potential of E. coli O157 cattle isolates

Nadejda Lupolova, Tim Dallman, Louise Matthews, James L. Bono, David L. Gally^*

^*Corresponding author for this work

Gastrointestinal Bacteria WGS

Research output: Contribution to journal › Article › peer-review

43 Citations (Scopus)

Abstract

Sequence analyses of pathogen genomes facilitate the tracking of disease outbreaks and allow relationships between strains to be reconstructed and virulence factors to be identified. However, these methods are generally used after an outbreak has happened. Here, we show that support vector machine analysis of bovine E. coli O157 isolate sequences can be applied to predict their zoonotic potential, identifying cattle strains more likely to be a serious threat to human health. Notably, only a minor subset (less than 10%) of bovine E. coli O157 isolates analyzed in our datasets were predicted to have the potential to cause human disease; this is despite the fact that the majority are within previously defined pathogenic lineages I or I/II and encode key virulence factors. The predictive capacity was retained when tested across datasets. The major differences between human and bovine E. coli O157 isolates were due to the relative abundances of hundreds of predicted prophage proteins. This finding has profound implications for public health management of disease because interventions in cattle, such a vaccination, can be targeted at herds carrying strains of high zoonotic potential. Machine-learning approaches should be applied broadly to further our understanding of pathogen biology.

Original language	English
Pages (from-to)	11312-11317
Number of pages	6
Journal	Proceedings of the National Academy of Sciences of the United States of America
Volume	113
Issue number	40
DOIs	https://doi.org/10.1073/pnas.1606567113
Publication status	Published - 4 Oct 2016

Bibliographical note

Funding Information:
We would like to acknowledge the value of human and bovine E. coli O157 sequence data available from previous published studies, especially work from the Wellcome Trust IPRAVE consortium, Public Health England, and the Scottish E. coli reference laboratory. This work was supported by Food Standards Scotland and the Food Standards Agency Grant FS101055 (to D.L.G., T.J.D., and L.M.), which has allowed the continuation of significant EHEC O157 research in the UK. This research was also supported by a University of Edinburgh studentship (N.L.) and core Biotechnology and Biological Sciences Research Council strategic programme Grant BB/J004227/1 (to D.L.G.). T.J.D. was funded by the National Institute for Health Research Health Protection Research Unit in Gastrointestinal Infections at the University of Liverpool in partnership with Public Health England, University of East Anglia, University of Oxford, and the Institute of Food Research.

Publisher Copyright:
© 2016, National Academy of Sciences. All rights reserved.

Keywords

Cattle
E. coli
Machine learning
Shiga toxin
Zoonosis

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1073/pnas.1606567113

Cite this

@article{a57ea7ba6d7342b68939c9d526183a31,

title = "Support vector machine applied to predict the zoonotic potential of E. coli O157 cattle isolates",

abstract = "Sequence analyses of pathogen genomes facilitate the tracking of disease outbreaks and allow relationships between strains to be reconstructed and virulence factors to be identified. However, these methods are generally used after an outbreak has happened. Here, we show that support vector machine analysis of bovine E. coli O157 isolate sequences can be applied to predict their zoonotic potential, identifying cattle strains more likely to be a serious threat to human health. Notably, only a minor subset (less than 10%) of bovine E. coli O157 isolates analyzed in our datasets were predicted to have the potential to cause human disease; this is despite the fact that the majority are within previously defined pathogenic lineages I or I/II and encode key virulence factors. The predictive capacity was retained when tested across datasets. The major differences between human and bovine E. coli O157 isolates were due to the relative abundances of hundreds of predicted prophage proteins. This finding has profound implications for public health management of disease because interventions in cattle, such a vaccination, can be targeted at herds carrying strains of high zoonotic potential. Machine-learning approaches should be applied broadly to further our understanding of pathogen biology.",

keywords = "Cattle, E. coli, Machine learning, Shiga toxin, Zoonosis",

author = "Nadejda Lupolova and Tim Dallman and Louise Matthews and Bono, {James L.} and Gally, {David L.}",

note = "Funding Information: We would like to acknowledge the value of human and bovine E. coli O157 sequence data available from previous published studies, especially work from the Wellcome Trust IPRAVE consortium, Public Health England, and the Scottish E. coli reference laboratory. This work was supported by Food Standards Scotland and the Food Standards Agency Grant FS101055 (to D.L.G., T.J.D., and L.M.), which has allowed the continuation of significant EHEC O157 research in the UK. This research was also supported by a University of Edinburgh studentship (N.L.) and core Biotechnology and Biological Sciences Research Council strategic programme Grant BB/J004227/1 (to D.L.G.). T.J.D. was funded by the National Institute for Health Research Health Protection Research Unit in Gastrointestinal Infections at the University of Liverpool in partnership with Public Health England, University of East Anglia, University of Oxford, and the Institute of Food Research. Publisher Copyright: {\textcopyright} 2016, National Academy of Sciences. All rights reserved.",

year = "2016",

month = oct,

day = "4",

doi = "10.1073/pnas.1606567113",

language = "English",

volume = "113",

pages = "11312--11317",

journal = "Proceedings of the National Academy of Sciences of the United States of America",

issn = "0027-8424",

publisher = "National Academy of Sciences",

number = "40",

}

TY - JOUR

T1 - Support vector machine applied to predict the zoonotic potential of E. coli O157 cattle isolates

AU - Lupolova, Nadejda

AU - Dallman, Tim

AU - Matthews, Louise

AU - Bono, James L.

AU - Gally, David L.

N1 - Funding Information: We would like to acknowledge the value of human and bovine E. coli O157 sequence data available from previous published studies, especially work from the Wellcome Trust IPRAVE consortium, Public Health England, and the Scottish E. coli reference laboratory. This work was supported by Food Standards Scotland and the Food Standards Agency Grant FS101055 (to D.L.G., T.J.D., and L.M.), which has allowed the continuation of significant EHEC O157 research in the UK. This research was also supported by a University of Edinburgh studentship (N.L.) and core Biotechnology and Biological Sciences Research Council strategic programme Grant BB/J004227/1 (to D.L.G.). T.J.D. was funded by the National Institute for Health Research Health Protection Research Unit in Gastrointestinal Infections at the University of Liverpool in partnership with Public Health England, University of East Anglia, University of Oxford, and the Institute of Food Research. Publisher Copyright: © 2016, National Academy of Sciences. All rights reserved.

PY - 2016/10/4

Y1 - 2016/10/4

N2 - Sequence analyses of pathogen genomes facilitate the tracking of disease outbreaks and allow relationships between strains to be reconstructed and virulence factors to be identified. However, these methods are generally used after an outbreak has happened. Here, we show that support vector machine analysis of bovine E. coli O157 isolate sequences can be applied to predict their zoonotic potential, identifying cattle strains more likely to be a serious threat to human health. Notably, only a minor subset (less than 10%) of bovine E. coli O157 isolates analyzed in our datasets were predicted to have the potential to cause human disease; this is despite the fact that the majority are within previously defined pathogenic lineages I or I/II and encode key virulence factors. The predictive capacity was retained when tested across datasets. The major differences between human and bovine E. coli O157 isolates were due to the relative abundances of hundreds of predicted prophage proteins. This finding has profound implications for public health management of disease because interventions in cattle, such a vaccination, can be targeted at herds carrying strains of high zoonotic potential. Machine-learning approaches should be applied broadly to further our understanding of pathogen biology.

AB - Sequence analyses of pathogen genomes facilitate the tracking of disease outbreaks and allow relationships between strains to be reconstructed and virulence factors to be identified. However, these methods are generally used after an outbreak has happened. Here, we show that support vector machine analysis of bovine E. coli O157 isolate sequences can be applied to predict their zoonotic potential, identifying cattle strains more likely to be a serious threat to human health. Notably, only a minor subset (less than 10%) of bovine E. coli O157 isolates analyzed in our datasets were predicted to have the potential to cause human disease; this is despite the fact that the majority are within previously defined pathogenic lineages I or I/II and encode key virulence factors. The predictive capacity was retained when tested across datasets. The major differences between human and bovine E. coli O157 isolates were due to the relative abundances of hundreds of predicted prophage proteins. This finding has profound implications for public health management of disease because interventions in cattle, such a vaccination, can be targeted at herds carrying strains of high zoonotic potential. Machine-learning approaches should be applied broadly to further our understanding of pathogen biology.

KW - Cattle

KW - E. coli

KW - Machine learning

KW - Shiga toxin

KW - Zoonosis

UR - http://www.scopus.com/inward/record.url?scp=84989829158&partnerID=8YFLogxK

U2 - 10.1073/pnas.1606567113

DO - 10.1073/pnas.1606567113

M3 - Article

C2 - 27647883

AN - SCOPUS:84989829158

VL - 113

SP - 11312

EP - 11317

JO - Proceedings of the National Academy of Sciences of the United States of America

JF - Proceedings of the National Academy of Sciences of the United States of America

SN - 0027-8424

IS - 40

ER -

Support vector machine applied to predict the zoonotic potential of E. coli O157 cattle isolates

Abstract

Bibliographical note

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this