Summary of our favorite PAG 2019 talks
February 13 2019
If you haven't already noticed, NCGAS really enjoys the Plant and Animal Genome (PAG) Conference that takes place every year in San Diego in January. Since we are from Bloomington, Indiana it's also a great time to leave behind the midwest winter and enjoy California sunshine for a week :).
We do realize this is a pretty expensive conference, and it is difficult for a lot of our community to attend. So in order to make the meeting more accessible, this year we presented research with our collaborators (link here), and plan to digest everything we learned and share the information with the community. Now, as promised in our previous blog post on how well we did at PAG 2019 (link here), below is a summary of some of the really interesting talks with links to abstracts, so you can follow up if you are interested!
- Going Full Circle: Assembly of High-Quality, Single-Contig Microbial Genomes from the Rumen Microbiome using Long-Read Sequencing (Abstract number PE1146)
Filling the Void: Assembly of High-Quality Microbial Genomes from the Rumen Microbiome using Short- and Long-Read Sequencing (Abstract number W0180)
This project work focuses on rumen microbiome. Samples were collected and sequenced using nanopore sequencing technology, the presentation walked through the steps taken to assemble and annotate the data, which was great notes for me. On the other hand, the presenter also had bandage plots of the assembly to show that they were able to generate 3 complete circles (indicative of complete genomes) and several other almost full circles (near-complete genomes). Through this analysis, they were now able to add three complete uncultured species to a bacterial database using long read technology with assembly only, no binning at all. They also have a research paper published on the data, available here.
I should mention there are more talks on microbiome in this session "The Analysis and Role of the Microbiome" which we always attend. Here is the link to this session to lookup other talks.
Rapid, Culture-Free Microbiome Deconvolution and Mobile Element Tracking using Proximity Ligation Technology (Abstract Number W1082)
Microbiomes with Hi-C. I liked this last year, and this talk wasn’t much different. But since I haven’t gotten anyone to try it yet: Hi-C lets you know what DNA is next to what, in a cell or nucleus. For bacteria, you can tell which cell contains which plasmid; for a protist, you know what the set of chromosomes in a cell is. This seems most attractive to me when you are doing metagenomes of a protist community—it’s pretty easy to filter out the bulk of the bacteria, and go from there on the protists.
Lessons Learned from a Naturally Evolved System to Enrich a Functional Microbiome: The Wonders of Mammalian Milk (Abstract number W886)
This talk, settled in the poultry genomics section for no apparent reason, was still one of the highlights of my PAG experience this year. The composition of mammalian milk and the changes it induces in the microbiome of humans during nursing is put forward in this talk to generally open up a dialog in all farm animals - what's normal for our development? Microbiome? What is normal for a domestic animal - a species that has been out of reach of natural selection for thousands of years? My own take - we don't often even recommend obligate carnivore diets for our dogs; dogs are not wolves and their digestive systems have evolved to better harmonize with their human companions. Maybe we should look a little more closely at our assumptions of a healthy human baby with a well-colonized gut full of Bifida
infantis, and maybe we should look more at what our chickens are eating too.
Genome assembly talks
Hi-C Assembly and Reassembly at the DNA Zoo: Inexpensive, Accurate and Open-Source (Abstract number W095)
The two points for me were that Hi-C data hasn’t been used to its full potential, and if it is used, all you need to get a reference genome is Illumina and Hi-C data. What this means is that all those awful Illumina assemblies of 3, 4, 5 years ago just need some Hi-C added, and that effort will be rescued.
Chromosome-Length Haplotigs for Cattle and Yak from Trio-Binning Assembly of an F1 Hybrid (Abstract number W177)
canuis becoming a hot topic these days! For those of you who haven't encountered it yet, Trio binning is a relatively new experimental design to make high-quality genomes from diploid organisms and in particular highly heterogeneous ones. One of the largest contributors to misassembly/fragmentation in genome assemblies, especially when using PacBio data, is haplotype difference. These differences can be structural or sequence-based, leading to breaks in the continuity of the assembly.
Trio binning addresses this in the experimental design - you take two animals (in this case a domestic cattle breed and a water buffalo) that have high heterozygosity between them and create an F1 hybrid. These three animals are your trio! Light coverage of each of the parents allows for binning of the deep coverage reads from the hybrid - allowing you to split the hybrid into "cow" and "water buffalo" reads, which are by default haploid. Each of these is assembled separately - giving you two very high-quality genomes.
This can be done within species as well - it was originally done with an Angus and a Brahman cow, and the speakers suggested that it could even be done in humans. Very cool stuff -
canueven ships with the ability to do trio binning now! Original trio paper here
Genome Assembly Gone Wrong: Lessons from the Vertebrate Genome Project (VGP) (Abstract number W017)
The VGP is a consortium we have mentioned before (in the Tegu Blog Post), so I went to their talk on what they have learned from working on numerous and diverse vertebrate genomes. No surprise, they are also gravitating toward Trio binning, but they also had other interesting insights. One, sadly, was that amphibians are just really difficult to do at this point, with their high repetition and sometimes (in salamanders) giant genomes. For those of us in the amphibian research world, we will likely be the last of the vertebrate genomes to have high-quality assemblies, as the needed technology/methods are still in need of more development.
Stem cell studies
Single-Cell Gene Expression from Arabidopsis Roots (Abstract number W486)
In this talk, the stem cells of the developing root meristem were separated. The most interesting part of the talk for me was the precision with which the different root cells were analyzed and categorized. I had no idea there were so many different cells in each developing root.
Stem Cells in the Northern White Rhino: Developing a Toolbox for Saving Species (Abstract number W918 )
Transcriptomic Analysis of Pluripotency in Northern White Rhinoceros Stem Cells (Abstract number W919)
I have been going to the rhino talks for several years now since I have a background in zoo biology. This is a VERY non-model species, but one that cannot wait for the development of tools to be vetted and tested. There are only two Northern White Rhinos left in the world - a mother and a daughter. With no males left, one would expect this sad story to be over. However, this dedicated group put in an amazing amount of work to make a legitimate plan to save this species!
Thanks to the frozen zoo, which has banked skin samples from a dozen northern white rhinos (starting in the 1970s - a serious forethought!). Building on the 2016 paper on induced pluripotent stem cells (iPSCs) from skin fibroblasts (link to paper here), the team was able to adapt the technology to rhinos. They talked about their steps in detail, walking through all the considerations in this extremely fast-paced talk. The second talk was even more amazing, as they walked through convincing themselves that these were legitimate iPSCs, that they were functional and pluripotent. Inducing gametes from iPSCs is extremely new technology, and only ever done in mice. But since rhino methylation timing is more similar to human than a mouse, the two groups are working together to produce sperm cells from the iPSCs.
I was blown away by how fast this group was able to build on revolutionary new technology in a dedicated mission to save the Northern White Rhino. Incredible stuff from San Diego Zoo.
Demography or Selection on Linked Cultural Traits or Genes? Investigating the Driver of Low mtDNA Diversity in the Sperm Whale using complementary Mitochondrial and Nuclear Genome Analyses (Abstract number W847)
This talk was on the global examination of sperm whale population structure; the title includes “selection on linked cultural traits”. Many of the talks in the Population and Conservation Genomics sessions consider the effects of past glaciations, and the suggested effects of global warming—here it was the Eemian glaciation 125,000 years ago.
Bulk-Bone Metabarcoding: Rapid Insights into Past Biodiversity and Ancient DNA Preservation (Abstract number W319)
Many archeological sites have relatively few intact bones, of either the humans or their food items. But there are huge piles of very small bone fragments that have been sieved out and put aside. If one takes this “bulk” bone, extracts DNA, and uses barcoding, one can learn a great deal about faunal assemblies and what food sources the people were using. I was intrigued by how much small pray were being taken, small rodents, lizards, etc.—where you would never find a big old thigh bone!
This talk was in the session Degraded DNA and Paleogenomics that is usually just fun: goat domestication, population history of the black rat, Gombe chimpanzees.
Pigeonomics: Using Whole Genome Data to Uncover the Genetic Bases of Morphological Variation in Domestic Pigeons (Abstract number W739)
Morphological changes in domestic animals, such as pigeons, are the result of years of selective breeding towards extreme phenotypes. Tiny snouts, such as in pug dogs, is also a trait sought after for pigeons. This snub beak is so extreme in some varieties that they cannot adequately feed their young. In this talk, the developmental genes responsible for these traits, as well as feathering of the feet, were explored.
Returning to the Sea and Beating Cancer: Comparative Genomics of Whales and the Evolution of Cancer Suppression (Abstract number W553)
This talk investigates Peto’s paradox (larger animals with more number of cells do not get cancer) in whale genomes. This study focused on the humpback whale which has shown that there are a lot of gene duplications in cancer-related genes and pathways suggesting that like elephants, damaged cells are not repaired but likely killed to prevent cancer. This talk did also make me aware of a fun fact - there are 12 reference whale genomes deposited to the NCBI.
Evolutionarily Informed Deep Learning Methods: Predicting Relative Transcript Abundance from DNA Sequence (Abstract number W488)
I was very impressed not only by the results and research question that this talk illuminated but by the delivery of the talk. I found that, even as a person with an informatics background, the talk was able to capture fine points of what machine learning is and how it works, while keeping it easy to understand for the wide audience. I want to hear this speaker again!
Exploring a Landscape of Genetic Variation in Virtual Reality (Abstract number C26)
Would you like to visualize your data in 3D! How much more fun would that be! This team from DNAnexus built a virtual reality tool call BigTop to explore Manhattan plots, Circos plots, and genome-wide association study (GWAS) results. This is an open source tool available on GitHub with datasets to visualize. Can’t wait to play with the datasets and the tool.
The list can go on but then there would be little difference between the PAG list and ours. If you would like to look up other talks, here is a link to all of PAG talks.
Stay tuned for an update on the third generation sequencing blog post as well!