Joint estimation of epidemiological and genetic processes for Mycobacterium bovis transmission dynamics in cattle and badgers

Lead Research Organisation: University of Edinburgh
Department Name: Royal (Dick) School of Veterinary Scienc

Abstract

The control and eradication of infectious diseases can be difficult for pathogens that are able to persist in multiple host species. This is the case for bovine tuberculosis (bTB), a disease primarily affecting cattle but also found in a number of wildlife species; in Britain and Ireland, the most important of these is the Eurasian badger (Meles meles). While Ireland has had a persistent bTB problem in cattle, by the 1970's bTB had been almost eradicated from Great Britain but since then the has been a dramatically re-emerging disease in cattle. BTB is a zoonosis with implications for both human and animal health, though chronic cases of either in Britain and Ireland are few. Control of bTB also places a severe strain on individual farmers, the farming industry and government, with a projected cost in England and Wales alone of over £1bn over the next decade. While it has long been suspected that badgers are involved, research efforts to date have not determined the extent to which badgers are responsible for eradicating bTB from cattle, and this topic is the subject of great social and political controversy.
One of the most important developments in epidemiology of the last few decades has been the increased use of 'genetic fingerprinting' to identify patterns of disease spread. Until recently, this has largely been done using only a small number of selected regions in the genome. While this kind of "genetic fingerprinting" has been very useful and shows that cattle and badgers in the same region are usually infected by the same bTB strain, the fingerprints are far from unique: many cattle and many badgers share the same type, making it impossible to determine who infected whom. In this project, we will take advantage of novel technology making it feasible and affordable to sequence the entire M. bovis genome for large numbers of samples. Because the bacterium occasionally makes mistakes while replicating its genome, new mutations constantly arise not seen using traditional fingerprinting methods but with the new technology creating a much more unique and discriminatory genetic fingerprint of transmission. Using samples collected over decades from cattle and badgers in Great Britain and Northern Ireland, we will sequence the genomes of hundreds of isolates to genetically track the spread of the pathogen and to test whether it is predominantly maintained in cattle, in badgers, or both. The unique opportunity exploited in this proposal is the availability of extraordinarily dense sampling of cattle and badgers infection together with entire life histories of individual cattle, including movement to other farms and whether it became infected with bTB at some point of its life. This creates an exceptional resource, allowing us to compare our very detailed understanding of contacts between cattle and between herds with the genetic fingerprint information. Based on this information, we will use mathematical models linked directly to statistical inference methods to simulate how the infection may have spread through cattle populations in Britain and Ireland and how it may have genetically changed in the process. This will be done under various different assumptions about the multiple possible sources and mechanisms of infection. By comparing our simulated results to the actual observations (e.g. the number of infected cattle and the type of bTB they carry, etc), we will gain unprecedented insight into the drivers for the spread of the disease and what may prevent its current control.

Technical Summary

Quantifying the transmission dynamics of multi-host pathogens is challenging, especially when sampling is biased. An important example of this is Mycobacterium bovis, the causative agent of bovine TB (bTB) an important disease of British and Irish cattle. M. bovis circulates in the badger and cattle populations and is currently expanding rapidly in GB. While it is well known that persistence of bTB in cattle is spatially localised, the critical question of how these spatial 'patches' are spread and maintained and the role badgers play are as yet unresolved questions. Here, we propose to combine whole genome sequencing (WGS) technology with detailed population data on both hosts to shed new light onto this problem. To address this question, we shall use the exceptional datasets available to integrate analyses across two organisational scales - the 'patch' scale of local persistence and spread and the transmission at the individual level within cattle herds and badger social groups. First, the natural spatial scale at which M. bovis is circulating shall be explored, using community structure algorithms from social network analysis to parse densely sampled phylogenetic trees by the most relevant clade structures. Second, the persistence and spread of bTB at the cattle herd and badger social group scale shall be investigated by analysing detailed life history and infection data. Mathematical models will be used to infer parameter distributions using Bayesian approaches; using methods that have already been used to study a number of important infectious diseases, including bTB. As the integration of genetic and epidemiological data ('phylodynamics') is as yet little applied to bacterial pathogens, not only shall this project generate important insights into bTB epidemiology, it shall also be a useful exemplar to others interested in mycobacterial phylodynamics in general.

Planned Impact

Government stakeholders. Bovine tuberculosis (bTB) places a significant burden on national economies in Britain and Ireland with an estimated future cost of £1 billion over the next ten years. New insights or approaches generated by this project therefore have the potential to make a direct economic impact if they affect government policy and, in the long term, translate into more effective control strategies. In addition to monetary costs, the question of how to best control bTB, and in particular the use of badger culling, has become a hugely divisive, political issue. Any information that would help to place control strategies on more a science- and evidence-based footing, should therefore be valuable to policy makers. Generally, benefits from this project to the public sector may include

I) insights into mechanisms of persistence and spread of the disease;
II) Risk factors influencing the distribution, size and re-occurrence of bTB outbreaks on farms;
III) evaluation of new molecular sequencing technology that could be incorporated into future diagnostic and control programs
IV) Identify likely mechanisms and ranges of spread of high incidence areas for bTB
V) Guidelines for bTB control, especially in terms of identifying methods to spatially target control and determine the extent of control required (both geographically, by species and by proportion of population)
V) Identification of unrecognised knowledge gaps and research priorities.

Cattle farming industry. Although there is legal compensation for cattle slaughtered as a result of testing positive for bTB, this covers only a fraction of the true cost, with a significant burden placed on farmers via the imposition of movement restrictions and the cost and effort associated with the multiple whole herd tests while the outbreak is being eradicated. This project will address long-standing questions about bTB epidemiology, persistence mechanisms and herd risk factors. Where answers to these question can be used to improve management and control strategies, either through government policy or self-guided action, they will result in tangible benefits to the livestock industry.

Wildlife conservation. The debate over the value of badger culling is an ongoing one, with sharply divided opinions on both sides, however it is currently considered one of the most important options for control of bTB. Better understanding of the role of badgers in the persistence of bTB will result in, at the very least, a more efficient approach to culling and/or provide better insight into the possible uses for vaccination.

Wider public. A large part of the public is generally familiar with the issues surrounding bTB, the cost it places on tax payers and farming communities and the debate about whether badger culling is a defendable control strategy on both practical and ethical grounds. By providing objective scientific information about some of the underlying processes, this project therefore has the potential to help shape public opinion on these issues. We would also hope that by demonstrating the value of using novel science and technology tools to answer questions of significant public interest, this work could play a part in highlighting the benefits of research, and the use of public funds, to society.

Academic Impact. For expected academic impact, see 'Academic beneficiaries'

Publications

10 25 50

 
Description This award is a continuation of a previous award at the University of Glasgow and thus the Key findings replicates the text associated with that award.

a key outcome of this project was the first analysis of bacterial sequence data from spatio-temporally connected cattle and badger populations. We analyses this dataset for evidence of transmission between and within species, and were able to estimates rates at which these transitions occurred. Also we were able to show that spatial proximity was the most important factor underlying close genetic relationships between samples taken from infected individuals, though time between samples and the network of recorded contacts (cattle movements between farms, and badgers moving between social groups) was also important.
Exploitation Route We have shown evidence of clear genetic signatures in the phylogenetic relationships between cattle and badger derived bacterial isolates for a particular epidemiological scenario, and identified analytical methods (a combination of machine learning analysis relating genetic data to epidemiology, and evolutionary analysis in a Bayesian inference framework - BASTA) that could be used to quantify the relationships elsewhere.

While we do not expect that same signature to be replicated everywhere, due to the dependencies on the epidemiological circumstances of any given scenario (e.g. spatial relationships amongst cattle and badgers, densities of the two species, availability of suitable contact points, environmental differences etc.) this finding provides confidence that obtaining similar datasets would similarly allow for quantification of the relationships between the infection dynamics in the two species. Importantly, this better understanding should allow for better targeting of control measures aimed at both species, and via ongoing analysis, allow for estimation of the efficacy of control, so long as a sufficiently long time series of data is available.
Sectors Agriculture, Food and Drink

URL https://www.thenakedscientists.com/articles/interviews/bovine-tb-spread-badgers-cows
 
Description This award is a continuation of an award held at the University of Glasgow - thus the text below replicates the text for that previous award. I) In the course of this award, a press conference was arranged by the Science Media Centre (SMC) in London on 17.12.2019 to highlight the results of this project that at the time were due to be published in the journal eLife. Initially, there was some concern about one of my public servant co-authors (Dr. R.S. Delahay) participating in the press briefing so soon after the election. However, in part due to an intervention by Defra's Chief Scientist, this was re-considered and Dr. Delahay was given permission to participate. The press briefing was (in my understanding) broadly viewed as successful and resulted in an accurate portrayal of the research and its significance). This outcome led to Defra and the SMC to meet to discuss "Lessons Learned" from the briefing. II) Part of the data generated in the course of this award related to the ongoing 'Test-Vaccinate-Removal' study area (https://www.daera-ni.gov.uk/articles/test-and-vaccinate-or-remove-tvr-wildlife-intervention-research), a study which aims to inform the future of bovine Tuberculosis control in Northern Ireland. Analysis of this work is yet to be completed (the TVR study itself only completed at the end of 2018) but our data and analysis have allowed for the direct quantification of the relative roles of cattle and badgers in this area, using methods closely related to those we have recently published in eLife. III) Results from this study were submitted as part of the consultation being done in the recent review of the scientific evidence for bovine Tb strategy, completed in Nov 2018 and led by Prof. Charles Godfray. This evidence contributed to the recommendation for future use of whole genome sequencing in evaluating Tb outbreaks and for its use in forensic examination of outbreaks.
Sector Agriculture, Food and Drink
Impact Types Societal,Policy & public services

 
Description Member of the Bovine TB partnership
Geographic Reach National 
Policy Influence Type Participation in a guidance/advisory committee
Impact The Bovine TB partnerships provides strategic advice on all aspect of bovine TB related policy. Bovine TB is a disease with economic and animal health impact, and affects farmers mental health and well being.
URL https://www.gov.uk/government/groups/bovine-tuberculosis-partnership
 
Description Submitted Contribution to: "A strategy for achieving Bovine Tuberculosis Free Status for England: 2018 review"
Geographic Reach National 
Policy Influence Type Implementation circular/rapid advice/letter to e.g. Ministry of Health
URL https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/7569...
 
Title Bioinformatics tool for identification of homoplasies in deep sequence data 
Description An analytical tool for identification of homoplasies was developed and made publicly available via github. A paper describing this method has been published. 
Type Of Material Improvements to research infrastructure 
Year Produced 2018 
Provided To Others? Yes  
Impact Too early to say. 
 
Title Homoplasy finder 
Description A freely available software tool to identify homoplasies in sequence data. 
Type Of Material Improvements to research infrastructure 
Year Produced 2019 
Provided To Others? Yes  
Impact Freely available tool to identify and annotate homoplasies on a phylogeny and sequence alignment. downloaded 105x as of 01/03/2020. 
URL https://github.com/JosephCrispell/homoplasyFinder
 
Title Bioinformatics tool for identification of homoplasies in deep sequence data 
Description An analytical tool for identification of homoplasies was developed and made publicly available via github. A paper describing this method has been published. 
Type Of Material Computer model/algorithm 
Year Produced 2018 
Provided To Others? Yes  
Impact too early to say 
 
Title Whole Genome Sequenced Mycobacterium bovis isolates sourced from infected cattle and badgers in England 
Description All WGS data used fin Crispell et al. in eLife 2019, have been uploaded to the National Centre for Biotechnology Information Short Read Archive (NCBI-SRA: PRJNA523164). Because of the sensitivity of the associated metadata, only the sampling date and species will be provided with these sequences. 
Type Of Material Database/Collection of data 
Year Produced 2019 
Provided To Others? Yes  
Impact Inclusion of these data in the M. bovis world genome project led by Dr. S Robbe-Austermann (USDA) 
URL https://www.ncbi.nlm.nih.gov/bioproject/PRJNA523164/
 
Description "Big Picture" article in the Veterinary Record 
Form Of Engagement Activity A magazine, newsletter or online publication
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact An article in the Veterinary Record, which is the journal associated with the British Veterinary Association, and therefore the primary communication medium for engagement with the Veterinary profession in the UK and also to farmers.
Year(s) Of Engagement Activity 2020
 
Description Articles in multiple newspapers 
Form Of Engagement Activity A press release, press conference or response to a media enquiry/interview
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Media (as a channel to the public)
Results and Impact Multiple news articles (in print and online - URL from the Guardian included as an example) on Crispell et al. 2019 in eLife, on the roles of cattle and badgers in the spread of bovine Tuberculosis. This coverage included:

The Guardian
The Telegraph
Independent
Daily Mail
iNews
Farming UK
Farmers Weekly
Stackyard
Farmers Guardian
Food & Farming Futures
Farming Life
Herald Planet
Irish Times
Go Tech Daily
Year(s) Of Engagement Activity 2019
URL https://www.theguardian.com/uk-news/2019/dec/17/tb-infection-from-cow-to-cow-more-likely-than-transm...
 
Description Blog post on "The Naked Scientist" 
Form Of Engagement Activity A broadcast e.g. TV/radio/film/podcast (other than news/press)
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Public/other audiences
Results and Impact The Naked Scientist is an internationally supported, national and international award winning podcast hosted by BBC East with supporting stations in Australia, NZ and South Africa. I was interviewed by the lead presenter about our work on whole genome sequencing of M. Bovis to determine the rates at which badgers and cattle infect each other, and in particular the results of the paper by Crispell et al. in eLife (2019).
Year(s) Of Engagement Activity 2019
URL https://www.thenakedscientists.com/articles/interviews/bovine-tb-spread-badgers-cows
 
Description British Ecological Society post - The Applied Ecologist: "One-size does not fit all: Insights from a novel outbreak of Mycobacterium bovis in Northern England" 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Public/other audiences
Results and Impact The blog post highlighted the results of a recent paper in the Journal of Applied Ecology, highlighting its research findings regarding the source and spread of a outbreak of bovine Tuberculosis in cattle and badgers, in an area where sustained infection had not previously been seen. the article highlighted how ecological settings influenced the manner of emergence and spread, and the use of bacterial whole genome sequencing to elucidate this. it also highlighted the effectiveness of efforts to control bovine tuberculosis in the area.
Year(s) Of Engagement Activity 2021
URL https://appliedecologistsblog.com/2021/11/04/one-size-does-not-fit-all-insights-from-a-novel-outbrea...
 
Description Interview on BBC Radio Four "Today" Programme 
Form Of Engagement Activity A press release, press conference or response to a media enquiry/interview
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Media (as a channel to the public)
Results and Impact Interview on the BBC Radio 4 Today Programme on the results of Crispell et al. at 8.40 am on 17.12.19, discussing what it tells us about the roles of badgers and cattle in maintaining bovine Tuberculosis in cattle. This interview was noted by multiple individuals (pers. comm.) including requests to present this work at a variety of venues (including the Annual meeting of the Midlothian Farmers Association).
Year(s) Of Engagement Activity 2019
 
Description Presentation at Midlothian Farmers Annual Meeting. 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Industry/Business
Results and Impact Presentation of work on whole genome sequencing of M bovis to midlothian farmer's association at their annual meeting.
Year(s) Of Engagement Activity 2020
URL https://www.midlothianadvertiser.co.uk/news/people/annual-midlothian-farmers-meeting-1-5084004