Guest post: Human genetics is microbial genomics

DWilsonThis is a guest post by Danny Wilson from the University of Oxford. Danny was recently awarded a Wellcome Trust/Royal Society fellowship at the Nuffield Department of Medicine, and in this post he tells us why you cannot understand human genetics without studying the genetics of microbes. If you are a geneticist who finds this post interesting, he is currently hiring.

Never mind about sequencing your own genome. Only 10% of cells on your “human” body are human anyway, the rest are microbial. And their genomes are far more interesting.

For one thing, there’s a whole ecosystem out there, made up of many species. Typically a person harbours 1,000 or more different species in their gut alone. For another, a person’s health is to a large part determined by the microbes that live on their body, whether that be as part of a long-term commensal relationship or an acute pathogenic interaction.

With 20% of the world’s deaths still attributable to infectious disease, the re-emergence of ancient pathogens driven by ever-increasing antibiotic resistance, and the UK’s 100K Genome Project– many of which will have to be genomes from patients (i.e. microbes) rather than patients’ own genomes given its budget – pathogen genomics is very much at the top of the agenda.

So what do pathogen genomes have to tell us?

  • Transmission. British researchers have been leading the use of genomics for routine “microbial forensics” in the NHS. Pioneering studies in Oxford and Cambridge focusing on the hospital-associated pathogens Staphylococcus aureus and Clostridium difficile have demonstrated how real-time genomics will revolutionize medical microbiology by detecting outbreaks as they occur.
  • Antibiotic resistance. Individual patients will benefit from pathogen genomics through accurate prediction of antibiotic resistance profiles. Already whole-genome sequencing outcompetes conventional techniques in speed and cost for some pathogens. High accuracy has been demonstrated in Escherichia coli and Klebsiella pneumoniae, pathogens in which the recent rise of antibiotic resistance has been particularly alarming.
  • Within-host evolution. For the first time genomics has detailed the microevolution of bacteria within the human body, revealing parallel evolution across patients in an outbreak of Burkholderia dolosa in cystic fibrosis patients, and an excess of protein-truncating mutations associated with disease progression in chronic Staphylococcus aureus carriage, underlining the potential of the body’s natural flora to evolve and ultimately subjugate its host.
  • Disease severity. Pinpointing the genetic basis of important microbial phenotypes – principal among which being disease severity – is the next big challenge for pathogen genomics. In other words, pathogen genome-wide association studies (pathogen GWAS). While much is known about the molecular mechanisms of virulence in microbes, population-based studies that ask why some bacteria make people more sick, or sick more often, are almost entirely lacking. By distinguishing hypervirulent from avirulent strains, and unearthing new avenues for drug design, pathogen GWAS has the potential to radically impact public health policy and individual patient management.

Important obstacles still lie in the way of genome-wide association studies in pathogens, because of fundamental differences between humans and microbes in the way they reproduce and exchange genetic material. But these challenges can be overcome through a detailed understanding of microbial population structure that whole genome sequencing is giving us.

Pathogen genomics is gathering pace – in the Modernising Medical Microbiology consortium alone we have sequenced over 15,000 bacterial genomes in the last 3 years – and ambitious projects such as 100K Foodborne Pathogen Genome Project at UC Davis aim to sequence many more. With such momentum, and innovation in our methods of analysis, it is only a matter of time before we begin making giant inroads into understanding the genetic architecture of infectious disease.

  • Digg
  • StumbleUpon
  • del.icio.us
  • Facebook
  • Twitter
  • Google Bookmarks
  • FriendFeed
  • Reddit

6 Responses to “Guest post: Human genetics is microbial genomics”


  • Completely agree. Could I add the EUFP7 PATHSEEK initiative to your links? The project is part of the translational arm of pathogen genomics, focussed on rapid clinical diagnostics and clinical management based on targeted enrichment of pathogen sequences from clinical samples. Understanding the bacteria is every bit as important as understanding the creature they live on/in!
    http://www.ucl.ac.uk/pathseek/project

  • Good article, I was wondering if you have a source for “Only 10% of cells on your “human” body are human anyway, the rest are microbial.” I have read this before (various %s), and would like to know were it comes from. Thank you

  • Yes I wondered about the 10% figure as well. If true, its because prokaryotic cells are smaller than eukaryotic cells. In terms of total cell volume its clear that we are mainly made up of our own cells, though bacteria still constitute a substantial fraction; and in terms of DNA, probably only a fraction is prokaryotic.

  • The short answer as to the source of the 10% figure appears to be: Luckey, TD (1972). Introduction to intestinal microecology. Am. J. Clin. Nutr. 25: 1292-94.

    A longer answer is in my blog post.

  • This is an open and exciting road to new understandings of disease, disorder and health.

    Mike, could you provide more details about the PATHSEEK program? What kinds of reports we will get? What is the current scale of the operation? Protocol testing? expanding? Who is qualified to be a member or partner?

  • Li Chen – More Pathseek info:
    The partners are fixed as part of the EU funding, but collaboration is always welcome. Most details on the website, which will be updated as the work proceeds. The project is about a year in and is currently finalising the universal capture methods together with the commercial partners. That should be reporting shortly. Then it will be about optimising speed and reliability, then testing and defining benchmarks for reporting. We hope to achieve widespread clinical adoption, which would generate datasets large enough to really help with progress on some of the key goals outlined by Danny in the post above.
    Get in touch for more information (project manager rachel.williams@ucl.ac.uk, or me, m.hubank@ucl.ac.uk).

Comments are currently closed.

Page optimized by WP Minify WordPress Plugin