Blog @ Illumina
Real scientists. Real commentary.

Microbes in the Mile-High City: Dispatch from the Discovery Symposium

Amy Cullinan Ph.D
| May 20, 2013

DenverThis year, Illumina hosted the first Discovery Symposium for Microbial Genomics prior to the American Society for Microbiology meeting in Denver, Colorado. On a breathtakingly sunny day in the mile-high city, nine exemplary researchers discussed how next-generation sequencing, like no other technology, has moved their fields of study forward.

First up, Jack Gilbert from Argonne National Laboratories spoke on a trio of projects that he and others are making contributions towards understanding the microbial dynamics of homes, hospitals, and planet Earth. Tweeting on his way up to the podium, Dr. Gilbert presented a fascinating view of ecological variability and structure of microbial communities. Considering all that our resident microbial communities provide for us, it’s fascinating to find evidence in mice and in humans that our microbial ecology fine tunes our very behavior. Speaking on citizen science projects such as the Home Microbiome study and American Gut, Dr. Gilbert emphasized that ordinary people can contribute to real data collection that will greatly expand our understanding of what modifies the microbiomes of our insides and our surroundings. Work on the Hospital Microbiome project has yielded data from 15,000 routine samples, suggesting that the bugs you pick up on your shoes after a trip to the clinic can persist for up to 7 hours afterward.  Finally, he introduced the Earth Microbiome Project, where abundant microbial taxa have been classified based on 16S rRNA sequencing, some of which may guide predictive modeling of potential carbon use scenarios. Jack concluded his talk with the sentiment that this is truly one of the most exciting times to be a microbial ecologist, and the tools we have access to now empower studies that were simply not possible even a few years ago.

Peter Evans, Chief of the Molecular Methods and Subtyping Branch in the FDA’s CFSAN division, which is responsible for developing methods for rapidly detecting and differentiating microbial foodborne pathogens spoke next about creating sequencing networks for molecular epidemiology. On the benefits of using sequencing-based subtyping to replace traditional bacterial typing schema, Dr. Evans spoke on the ability to perform rapid and accurate attribution, risk assessment, and modeling. Providing evidence from recent outbreaks of cannoli to cereal grain, he and others envision sequencing labs sharing whole-genome sequencing data and analyses in real time to detect and investigate the cause of foodborne outbreaks. Dr. Evans described several projects underway, including ones like the 100K Genome Project and GenomeTrakr would use next-generation benchtop sequencers like MiSeq to create a database for federal, local health agencies, researchers, and the general public.

Joerg Graf from the University of Connecticut presented a compelling comparison of metagenomics datasets that were generated on 454 GS-FLX compared to the MiSeq system. Working with different primers specific for regions of 16S as well as different sequencing platforms, Dr. Graf showed excellent data quality and higher coverage on the MiSeq run. Accompanied by wonderful visuals from termites, bobtailed squid, and medicinal leeches, Dr. Graf discussed de novo- and reference-based clustering to determine operational taxonomic units (OTUs),  that define the branchpoints of phylogenetic trees. Interestingly, He found the choice of sequencing technology did not have a significant effect on taxonomic composition comparisons from low to high complexity samples.

Clotilde Teiling, Senior Product Manager at Illumina, next gave us a preview of her ASM poster on sequencing 96 strains of Saccharomyces cerevisiae with HiSeq 2500. In collaboration with San Diego’s White Labs, they sequenced yeast strains involved in brewing beer in order to capture the biological diversity and gain insight into the difference between the strains. Might this tantalizing evidence explain why German and Belgian beers have truly different tastes?

Cheryl Tarr, Chief of the Listeria, Vibrio, Yersinia, and Enterobacteriaceae Reference Laboratories at the Centers for Disease Control and Prevention, spoke on whole genome sequencing and bioinformatics approaches to characterize emerging pathogens, in particular Vibrio cholerae. Excited that molecular data is finally being adopted for use in public health and microbial evolution studies, Dr.  Tarr told the fascinating story of how a molecular epidemiology approach supported the theory that the 2010 cholera outbreak in Haiti was the likely due to human-initiated activity, namely Nepalese aid workers. Using whole-genome sequencing of 60 vibrio strains, Dr. Tarr and others determined that gene content changes were not due to acquisition of pathogenicity islands, uncovering a faulty DNA uptake mechanism. A defect in this vital system for maintaining bacterial diversity is a hopeful sign that this particular scourge will have an early extinction.

Feng Chen from the Joint Genome Institute of the Department of Energy spoke next on high-throughput metagenome community characterization. Dr. Chen described his ideal pipeline for next-generation screening, in that it must accommodate a broad spectrum of samples, and the methodology should be applied to different regions of the 16S rRNA gene. The results must be extremely high quality, and the process streamlined for reproducibility. Using varied metagenomics examples from the fungi inside the rumen of a “windowed” cow to agave microclimates, Dr. Chen explained why MiSeq was his choice for developing such a pipeline, highlighting the improvements in the Real-Time calling algorithm (RTA) with a novel by-cycle phasing correction that decreases the need to spike in PhiX control for lower diversity samples.

Next up was a great interactive session from Jim Huntley, previously an Illumina Field Application Scientist, now of the University of Colorado at Boulder and the amazing BioFrontiers Institute. The smartly titled talk on “What to expect When You’re Expecting Your MiSeq” provided a wealth of very practical information for new users, and those who are considering instrument purchase. Jim covered operational aspects of MiSeq ownership from uninterrupted power supply to library QC methods, to training your technicians to perform runs, with many tips and tricks gleaned from his long instrument experience. Gearing examples toward microbial sequencing applications, Jim walked through setting up an example 16S rRNA project. His very useful slide deck, as well as several others can be found at the meeting app: (one-time login required)

Rob Knight from the BioFrontiers Institute, the Howard Hughes Medical Institute, and University of Colorado at Boulder presented next on several sweeping projects on Earth and human microbiome studies. Starting off with why you should care about your microbial community (sex, drugs, and how likely you are to get bitten by a mosquito, apparently), Dr. Knight described the utility of the Quantitative Insights into Microbial Ecology (QIIME) tool, an open source software package for comparison and analysis of microbial communities.  QIIME is rapidly becoming a the workhorse of choice in this field to transform read-level sequencing output through taxonomic identification, and construction of phylogenetic trees. You can also do some pretty neat statistical analysis and data visualization, and end up with publication-quality graphics. Dr. Knight gave us a look at some fascinating data on the microbiota of neonates after vaginal versus c-section delivery and the Earth Microbiome Project, highlighting the importance of using next-generation sequencing to expand the reach of such studies, building more robust datasets with greater numbers of samples.

To close out this riveting day of microbial genomics, associate director Niall Gormley from Illumina’s Chesterford research group gave a historical perspective on the evolution of sample prep methods, and how they have been optimized over the years since the debut of the original GA in 2007. Dr Gormley reminded us that the one guiding principle that remains behind sample prep development is keep it simple- in terms of transfers, clean up steps, and reasonable volumes. For Mate Pair protocol development, manipulating large molecules was a tricky step greatly improved by the addition of the Nextera chemistry, and Niall showed some very nice de novo assemblies.

The ASM Discovery Symposium was a truly inspirational group of microbial genomics success stories, and a great illustration of how next-generation sequencing is fueling a revolution.  A big thank you to all the speakers and attendees who participated! Check out our album on FaceBook for a look at the day's events, and where we all ended up watching the sunset from 5,280 feet above sea level.