Bruce Deagle

My research focuses on the application of advances in the field of genetics to better understand ecology and evolution of animal populations. My current position (2018) is as a Research Scientist leading the Ecological Genetics group at the Australian Antarctic Division (AAD). We are currently working on plankton DNA metabarcoding, penguin diet  and a range of other projects. For my PhD thesis I studied DNA-based methods for determining animal diet and continue to do work in this field. I have also studied the evolutionary genetics of threespine stickleback populations.

                                                            


Updates:

_____________________________________________

Sept 10, 2018 - A paper led by Laurence Clarke titled 'DNA-based diet analysis of mesopelagic fish from the southern Kerguelen Axis' has been published in  Deep Sea Research Part II: Topical Studies in Oceanography. In this study, mesopelagic fish (those living mostly in mid-water ~100-1000m of the open ocean) were dissected and food remains in their stomachs characterised using DNA metabarcoding. The samples were collected during the  2016 K-Axis research voyage to the Southern Indian Ocean and the paper is part of a special journal issue focussed on the findings of this broad ecosystem study. Laurence's work had a number of interesting findings: the Lanternfishes (Myctophidae) ate mostly amphipods, euphausiids and copepods, whereas the Deep-sea Smelt (Bathylagus sp.) ate many gelatinous animals, such as jellyfish; the diet of all the fish species changed with fish size (i.e. larger fish tended to have krill present in stomach). To examine a technical issue, we determined if secondary predation (i.e. prey from the stomachs of food items) would be detected by our method. This was done by removing relatively intact prey items from the fish stomach and processing them using the standard technique - any extra DNA signal not from the prey item itself is likely from their stomach contents. The signal from secondary predation was very weak in this experiment, indicating the fish directly consumed most of the species that we detected.


Figure  Myctophid fish sample with stomach being dissected for DNA analysis and plot showing relationship between fish size and detection of krill.

___________________________________________

June 22, 2018 - A paper I've been working on with a group of colleagues on DNA-based diet analysis has been published in Molecular Ecology (all the details are available here). The article summarises our thoughts on how to interpret counts of DNA sequences we get from samples in animal diet studies (usually faecal samples). It is relevant to the research we do on seabirds here at the AAD, but also in a wide range of other diet studies, and anywhere DNA sequencing is used to measure biodiversity.


Figure  Data in 10 faecal samples from different animals summarised using alternative metrics. Each vertical column represents a faecal samples, colours represent different food taxa detected with DNA markers. Data are from previously published research by some of the authors of the new synthesis paper: black-browed albatross (Julie McInnes), insectivorous bat (Eero Vesterinen) and Grevy’s zebra (Tyler Kartzinel).
____________________________________________

June 20, 2018 - Three posters from our group are being presented at a major conference for Antarctic/Polar science in Switzerland (POLAR 2018). Laurence Clarke is presenting his work on genetic detections of a parasite that he has shown is extremely common near the ice edge in the southern Indian Ocean - it is also found in oceans globally, and may have an important role in marine ecosystems. Laurence is also presenting our work on genetic monitoring of zooplankton (poster below). Claire Waluda, a colleague from the British Antarctic Survey, is presenting work we've done comparing penguin diet determined using faecal DNA versus stomach flushing (poster below). Click images for larger size.


       

_____________________________________________

June 18, 2018 - A nice story on the Australian Antarctic Division website covering the environmental DNA project we are working on can be found here. It includes rare footage of me in the lab, taking credit for the work actually being done by Andrea...

____________________________________________

March 27, 2018 - Andrea Polanowski and I from the AAD genetics group just came back from doing marine science on a short Aurora Australis voyage (V4 Macquarie Island resupply). We were collecting environmental DNA samples from sea water as well as zooplankton specimens (using the continuous plankton recorder) - these will be used to refine DNA-based methods of measuring marine biodiversity (see link to project here). We collected several hundred DNA samples from the open ocean between Tasmania and Macquarie Island as well as from transects close to the subantarctic island. The photo below shows a view from the Island with the ship Aurora Australis just visible on the horizon.  


_____________________________________________

December 21, 2017 - A paper by PhD student Ricardo De Paoli-Iseppi  has been published in PLOS ONE: DNA methylation levels in candidate genes associated with chronological age in mammals are not conserved in a long-lived seabird (PDF).

In this study Ric set out to find genetic markers that could be used to determine the age of seabirds. To do this he's been working with a population of short-tailed shearwaters which have been monitored for decades and the ages of most individuals are known (birds are identified by leg bands put on when they are chicks). In this paper he focussed on genes that have been shown to have age-specific chemical modifications in mammals (epigenetic changes - DNA methylation - observed in humans, mice and/or whales). His data show that these genes don't have the same patterns of epigenetic changes with age in shearwaters. The level of DNA methylation observed in birds is also often quite different compare to what has been observed in mammals. This suggests that the control of genes (regulated by DNA methylation) varies between different groups of organisms and this makes finding age-related markers in birds a bit harder. Ric is now moving on to a new approach which scans the genome of the shearwater to find age-related markers directly in his study species.

Figure  DNA methylation level measured in Mouse (left); from Spiers et al. 2016), and Shearwater (right) for a gene called MYOD1 at a conserved CpG site. The different colours show males and females. Note the difference in scale of the vertical axis, and some correlation with age in mice (this marker combined with several others allow age estimation in mice).

___________________________________________

December 5, 2017 - There are quite a few molecular ecologists based in Hobart and we use similar approaches in our research, but we are spread across different research institutes and study very different organisms and ecosystem, so don't often cross paths. To strengthen the links in this community and to hear some interesting science talks Laurence Clarke and I organised the Hobart Molecular Ecology Workshop. The program included 17 talks on a wide range of topics, From marine microbes and Eucalyptus adaptation genetics to shark populations and using molecular methods to trace infections in hospitals. Thanks to the ACE-CRC for providing snacks, lunch and a lovely venue on the Hobart waterfront.

_____________________________________________

November 22, 2017 - A paper we've been working on for quite a while has been accepted in Molecular Ecology Resources (ReadOnly PDF). The study looked at samples collected using the Continuous Plankton Recorder (CPR). This device has been used to characterise zooplankton biodiversity along transects covering hundreds of thousands of kilometres in the Southern Ocean CPR survey. Plankton collected in the survey is currently identified using classical taxonomy (i.e. using a microscope and morphological features). We investigated the potential to use DNA metabarcoding (species identification from DNA mixtures) as a tool for rapid collection of taxonomic data from CPR samples.


In our study, zooplankton were collected between Tasmania and Macquarie Island. Plankton were identified by microscope and by DNA sequencing (COI marker).
DNA increased the number of species identified and provided high resolution taxonomy of groups problematic in conventional surveys (e.g. larval echinoderms and hydrozoans).  Metabarcoding also generally produced more detections than microscopy, but this sensitivity may make cross-contamination a problem. In some samples the prevalence of DNA from large plankton (e.g. krill) masked the presence of smaller species (e.g. copepods). Overall, the genetic data represents a substantial shift in perspective, making direct integration into current long-term time-series challenging.

One of the main outcome of the work was an evaluation of how DNA methods can move from the current snapshot studies to the requirements of a long-term monitoring program. This topic has been discussed in a number of other recent paper (e.g. a nice study looking at freshwater ecosystems). The particular areas that we think need special consideration for standardised surveys include:

(1) Assigning taxonomy to DNA sequences This step that is still far from perfect. At the very least, an imperfect, but standard, method needs to be adopted. The sequence data can be reanalysed as the methods improve, which is a strength of DNA metabarcoding.  

(2) Contamination A well documented problem: miniscule amounts DNA can be recovered and used for biomonitoring, but trace DNA in laboratories or on sampling equipment will also be detected. Just need to be vigilant, hopefully should be less of an issue in long-term studies with large datasets. 

(3) Relative biomass versus presence-absence data Are the proportion of DNA sequence reads reflect of the relative abundance (biomass) of each species in a sample? Are detection sensitivities consistent across samples when we convert data to a presence-absence matrix?

(4) Estimating biomass with an internal standard To compare the relative amount of plankton DNA across samples we tried adding a standard amount of mouse DNA to each sample. This approach may have some merit, but needs further testing.

(5) Generating coherent long-term datasets.One of the great strengths of the CPR survey is that it has used essentially the same technology for the past 80 years. In contrast, high-throughput sequencing is recent technology in a state of flux making it hard to adopt a standard protocol.



Figure  (a) Map of sampling locations. (b) Bar plots illustrating the DNA marker sequences recovered from different marine organisms along each transect. (c) PCoA plots using binary Jaccard distances from DNA sequences (rarefied) and morphological datasets.

_____________________________________________

November 17, 2017 - Over the last few months I have had the opportunity to attend a couple great meetings. I gave a plenary talk at a Symposium on Ecological Networks and Molecular Analysis of Trophic Interactions in Sweden. It was fantastic to meet many scientist whose work I have been reading for many years and see many interesting applications using DNA in diet studies. Closer to home I also attended a Symposium on Kerguelen Plateau Marine Ecosystem and Fisheries this covered broad topics relevant to an amazing part of the Southern Ocean.
_____________________________________________

August 31, 2017 - Three papers have been published recently by PhD students Julie McInnes and Ricardo De Paoli-Iseppi who are both working in the AAD Ecological Genetics laboratory.

Julie's two papers are the last chapters from her thesis and use DNA to document the diet of black-browed albatross (and closely related Campbell albatross) throughout their circumpolar range. One paper is focused on the broad diet and highlights the prevalence of jellyfish. Her other paper focuses on the type of fish in the albatross diet and consumption of fisheries discards in some locations (e.g. Falkland Islands and Iles Kerguelen). More details are available here: 

McInnes JC et al. (2017) High occurrence of jellyfish predation by black-browed and Campbell albatross identified by DNA metabarcoding. Molecular Ecology 26:4831-4845

McInnes JC et al. (2017) DNA Metabarcoding as a Marine Conservation and Management Tool: A Circumpolar Examination of Fishery Discards in the Diet of Threatened Albatross. Frontiers in Marine Science: 4:277 
(PDF)

Figure Comparison between back-browed albatross fish prey and fishery catch amounts at the Falkland Islands by month from December 2013-March 2014 (excluding February) and October 2014 to March 2015. Solid borders represent New Island; dashed borders represent Steeple Jason. (A) Scats with or without fishery target species (black and gray bars, respectively), compared to the total catch in the fishery (blue line). 


Ricardo's paper
 is from the first chapter of his thesis - a review of work that has been done to use signals in the genomes of animals (DNA methylation) to determine their age. This is the topic of his PhD, a relatively new and exciting field!

De Paoli-Iseppi R et al. (2017) Measuring animal age with DNA methylation: from humans to wild animals. Frontiers in Genetics: 8:106 
(PDF)



Figure  Timeline of the major studies and tissues analysed for global or targeted DNA methylation in the review. Studies are age-associated except where indicated.
___________________________________________
May 15, 2017 - Enjoyed being involved in the seventh DNA metabarcoding Spring School in Portugal at the beginning of the month (see metabarcoding.org)! It was a great to hear about a large number of interesting projects and thoughts from both the instructors and students. 
_____________________________________________

Feb 22, 2017 Two papers have recently been published featuring work carried out in the Ecological Genetics group.

The first paper was lead by Laurence Clarke and looks at the performance of different DNA metabarcoding markers for comparing samples of zooplankton communities. This work was done in collaboration with marine ecologists who characterised parallel samples using morphological methods. Three DNA barcode markers were examined, the standard barcode of life marker (COI) did a good job at both recovering a wide range of different species and providing high resolution taxonomy due to a large reference database. The ability to recover DNA from a range of species with this marker was surprising given the large amount of variation in primer binding sites (see post below from Sept. 2014). The paper is available here 

                                 

Figure  Plankton net sampling and cluster diagram showing crustacean families detected with three metabarcoding markers.

The second paper describes work done characterising the Antarctic krill transcriptome (i.e. active genes within the krill's genome). This work was done in collaboration with researchers from the University of Padua, Italy - this team of bioinformaticians produced a 'master' transcriptome by collating past datasets with more than 360 million new Illumina sequence reads generated by the AAD from larval krill. The paper is available here and users can access this new resource in an online database at http://krilldb.bio.unipd.it/



Figure   The home page of KrillDB and an example of search results

___________________________________________

Jan 17, 2017 PhD student Julie McInnes recently published a paper looking at sample collection issues for DNA diet studies of vertebrates (see paper here). The work is based on results from shy albatross scat collections she made over a two year period. DNA data were collected from the samples using high-throughput Illumina sequencing of a marker which recovers information from all major animal lineages (V7 region of the nuclear small subunit ribosomal DNA gene [18S]). I think one of the most interesting things about the paper is it clearly shows that proportions of DNA from various sources (i.e. diet and non-diet components) in animal faeces varies between samples considerably . Sometimes dietary DNA is a relatively high proportion (~20%), other times DNA from diet items is only a tiny fraction of what is present. This variation in diet information is masked when using group-specific markers (e.g. recovering DNA from only particular prey groups) since these studies don't quantify the total amount of DNA. With group specific markers, only targeted taxa are recovered, therefore you can't differentiate between samples where diet sequences represent 0.5%  of the sequences versus 50%. Julie's broader investigation of what information is in scats has important implications about the quality of data we get from samples. Fortunately, Julie's data also show that we can predict which samples provide the highest quality data in dietary studies - this has practical implications for ongoing Antarctic seabird diet studies at the AAD and for other researchers using this approach. Julie put together a nice video summarising of some of the recommendations (video here).



Figure   Variation in proportion of food DNA in scats for each breeding stage (incubation, brood and chick rearing) and various distinct cohorts of birds within these stages. The best diet data are likely to come from samples taken from breeding birds during brood stage when the adult take shorter trips and return to colony often (Photo of Julie collecting samples taken by Kim Kliska).
_____________________________________________

Oct 24, 2016 - A nice application of some of the DNA-based methods for dietary analysis developed at the AAD has been published (see abstract here). This work was carried out by Austen Thomas and the study looks at Pacific salmon species being eaten by harbour seals in coastal waters of British Columbia, Canada (Strait of Georgia). This region is of interest because salmon migrating to and from local rivers (adult and juvenile salmon respectively) support important First Nation, commercial and recreational fisheries. In the study over 1000 scats were collected over two years (4 different collection locations). The bones and fish DNA in each sample were characterised. Bones from different salmon species are hard to distinguish, and those from small fish are are often completely digested; however, the DNA methods allowed the five different species of salmon to be distinguished and also detection of juvenile salmon consumption . Herring and gadoid species were most common fish consumed by the seals, but salmon constituted a significant proportion of the diet both during fall (mostly adults salmon returning to spawn) and spring (mostly juvenile salmon). The two salmon species of highest conservation concern (coho and Chinook) were not the major salmon prey consumed in the fall, but were being consumed more than other salmon in the spring as juveniles. 


Figure  Plots showing proportion of salmon species DNA sequences recovered from seal scat samples in  (density plots based on bootstrap replicates).

_____________________________________________

Sept 16, 2016 - A book chapter on the 'Genetics of Antarctic Krill' written by Simon Jarman and I has been published in a book on the 'Biology and Ecology of Antarctic Krill'. Other chapters include one on krill reproduction and development written by AAD scientist So Kawaguchi and one by former AAD scientist Steve Nicol and Jacqueline Foster on the krill fishery.


_____________________________________________

May 30, 2016 - A new paper has been published on Patagonian toothfish population genetics by Lola Toomey (Toomey et al. 2016) who worked as a student intern in our group last year. Lola used DNA extracted from otoliths to characterise variation in mitochondrial and nuclear DNA markers from locations throughout the Southern Ocean. One of the clear results is that the Pacific sector (represented by Macquarie Island samples) was genetically distinct from the Indian and Atlantic sectors of the Southern Ocean. The study also opens the way for other genetics studies to take advantage of the large otolith collections which are often maintained for aging commercially important fish species.

Figure   Network showing relationships between mitochondrial sequences (Ctr1, CytB3 and COI) for six locations: South Sandwich Islands (SSI), South Georgia (SG), Macquarie Island (MACCA), Kerguelen (n = 17), Heard and McDonald islands (HIMI; n = 97) and Crozet (n = 36). Each haplotype is represented by a circle, and coloured circles with a number indicate the presence of this haplotype in the population and its frequency. A non-coloured circle indicates that this haplotype is not present in the population concerned. A line separating two haplotypes represents a mutation that differentiates these haplotypes, any additional mutation is represented by a black dot. Lines that connect different layers indicate shared haplotypes between populations.

_____________________________________________

Jan 5, 2016 - Over the next few month I will be part of a research team studying/collecting samples from the Kerguelen Plateau region of the Southern Ocean (see map below). The overall project will be looking at physical, biological and chemical conditions across this area. My particular focus (along with Laurence Clarke) will be on documenting the patterns of genetic biodiversity in the plankton - this will involve collecting organisms of various size fractions from seawater during the voyage - everything from visible zooplankton to single-celled protists (see DNA barcoding plankton). The genetic information will provide a new perspective on the biodiversity of the region, and the collections will allow us to test new methods of efficiently characterising plankton communities -  enabling ongoing detailed ecological studies of Antarctic marine ecosystems.

Figure  The k-axis voyage will study the region north of Davis station including the southern part of the Kerguelen Plateau 

_____________________________________________

Dec 20, 2015 - A study carried out by recently graduated UBC PhD student, Austen Thomas, has been published in Molecular Ecology Resources (Thomas et al. 2016). The paper looks at optimizing the accuracy of DNA metabarcoding when using the approach for quantifying relative species abundances in biological samples. The idea is to correct for species-specific biases in recovery of DNA. This is done by preparing 50/50 mixtures of tissue from a target species and a control species, then sequence DNA extracted from the mixtures to generate relative correction factors (RCFs). These species-specific RCFs essentially measure how under- or over-represented a particular species was relative to the control tissue. The RCFs could be applied to sequence data from environmental samples (in Austen's case seal scat mixtures) to weight the sequences recovered from each species appropriately (i.e. under-represented species multiplied up, over-represented species multiplied down).


Figure  Steps involved in calculating relative correction factors (RCFs) from a prey tissue library (from
Thomas et al. 2016)

_____________________________________________


Oct 30, 2015 - Results from our study on Antarctic krill population genomics are now published in Molecular Ecology (Deagle et al. 2015). As mentioned in a previous post the aim of the work was to determine if there are genetically distinguishable "stocks" of krill in the waters around Antarctica. During the study we developed new krill genetic markers (RAD markers) to allow us to characterise genetic variability within and between sampling sites. Finding good RAD markers proved to be quite difficult due krill's large genome size and the fact that many sequences were present in multiple copies (see January post below). Using methods of data analysis appropriate for multi-copy variants, we detected genetic structure caused by individual and technical variability, but no population related structure. This conclusion was supported by a small number of higher quality RAD-seq loci and separate analysis of a more traditional marker (mtDNA).  The lack of population genetic structure of Antarctic krill throughout their distribution shows that, from a genetics perspective, krill from throughout the Southern Ocean are one large mixed population. As discussed in the paper, this conclusion needs to include the caveat that the process of genetic differentiation between krill populations will be quite slow due to species’s enormous population size. Therefore, lack of genetic differences between locations does not necessarily indicate that there is an ongoing high level of movement between regions - this is an important point in the context of fisheries management.

Figure  Lack of genetic differences between krill collected at four Southern Ocean locations based on analysis of  more than twelve thousand RAD markers. In this plot each point represents an individual krill and collections are identified by colour. The analysis (principal component analysis) attempts to find variation in this large dataset and will separate points accordingly; however, no consistent variation is found, and krill from different sites are not separated. This indicates that krill found thousands of kilometers apart are no more distinct than krill within a single location.

_____________________________________________

Sept 1, 2015 - A  paper reviewing molecular approaches for estimating animal age has been published in Molecular Ecology (Jarman et al. 2015). Simon Jarman is the lead author and this review follows a previous paper from the Ecological Genetics group at the Australian Antarctic Division looking at epigenetic markers for age in humpback whales (Polanowski et al. 2014). As a group we have tried a few different approaches to provide non-invasive age estimates from whale tissue samples and now with several new approaches available this area of research is showing much potential. So far most of the detailed work has been done in model organisms, but some nice studies on wild animals are being published. This review provides a useful summary of the current research and future directions.
Figure  Examples from the review paper showing promising molecular age biomarkers  (a) The relationship between age and methylation of three CpG sites in humpback whales (data from Polanowski et al. 2014) (b) Age estimation in mosquitoes based on multiple mRNA markers (data from Wang et al. 2013).

_____________________________________________

May 16, 2015 - Over the last few month we've been starting work on a new project on plankton DNA metabarcoding - essentially developing new genetic approaches for efficiently identifying the tiny organisms which play a key role in Southern Ocean ecology. This work is being done in collaboration with AAD scientists with expertise in the taxonomy and ecology of zooplankton and single-celled 'protists'. There is also a new molecular ecologist post-doc (Laurence Clarke) working on the project (he's funded through the Antarctic Climate & Ecosystems Cooperative Research Centre). We have just started analysing our first samples - some of which I collected in April on a voyage from Hobart to Macquarie Island (V4). The photo below from the voyage shows crew of the Aurora Australis recovering a Continuous Plankton Recorder which was towed behind the ship.

    


_____________________________________________

January 6, 2015 - In the Antarctic krill population genomics project that I've been working we've been trying to determine if there are genetically distinguishable "stocks" of krill in the waters around Antarctica. One key aspect of the project is the development of new krill genetic markers (RAD markers) to allow us to characterise genetic variability within and between sampling sites. However, producing a reliable RAD genotype dataset for krill has proved to be challenging. The issue is caused by the prevalence of repetitive DNA elements in the krills' massive genome. The RAD method allows a huge number of genetic markers to be isolated, and variation can be characterized by sequencing each of these markers many times in DNA samples. But problems arise if markers have been duplicated within the genome. For example, if a gene is duplicated on a chromosome to create FormA and FormB it is not always possible to distinguish variation variation in FormA from variation between duplicates (i.e. between FormA and FormB). This is especially difficult in genetically diverse species, such as krill, because distinguishing gene duplicates based on their divergence is not reliable.

One manifestation of having multiple copies of RAD markers is that when you obtain sequences from an individual marker in a particular sample, variable nucleotides aren’t recovered in the expected 50:50 ratio (i.e. one half of the sequences should come from each parent). Instead the ratios can be skewed because a particular nucleotide could be fixed in one copy of the duplicated marker and variable in the other copy (in this case you’d expect the ratio of recovered sequences to be 75:25). Below is a plot showing the expected versus observed counts of sequences from our dataset - these ratios are far from 50:50 and genotypes cannot be reliably called. This issue has lead us to pursue alternative methods of analysis.


Figure  (a) Plot of reference versus alternate nucleotide counts in sequences from a single krill. Each point represents a proposed biallelic SNP marker; green points show those scored as heterozygotes, red points show those loci scored as homozygous. The points circled in blue show markers with genotype calls that changed in a replicated sample processed with DNA from the same krill. Inset shows the expected distribution of counts from random sampling of binomial distribution with p=0.5.
_____________________________________________

September 16, 2014 - A short critical evaluation of COI as a DNA metabarcoding marker was just published as an opinion piece in Biology Letters (Deagle et al. 2014). It was inspired by the growing number of high-throughput sequencing studies that are using DNA barcode amplicons to characterise mixed DNA samples. In particular, it is a response to the suggestion that COI could be the default DNA metabarcoding marker in ecological studies of animals:

Because of [COI primer] success across a broad spectrum of metazoan phyla, we also envision their use in creating amplicon libraries from environmental samples such as planktonic communities or from gut contents for food web analyses… “ Geller et al.  2013 Mol. Ecol. Res.

“Unfortunately, most studies of environmental samples have targeted ribosomal markers, despite the fact that the mitochondrial Cytochrome c Oxidase subunit I gene (COI) is by far the most widely available sequence region in public reference libraries.” - Leray et al. 2013 Frontiers in Zoology

    The Barcode of Life initiative has produced a wide range of valuable scientific outcomes that wouldn't be possible without standardised markers. However, recovering DNA markers for species identification from DNA mixtures is not a straightforward application of standard DNA barcoding. The main issue is that major biases can be introduced due to differential recovery of DNA from component species (often caused by variation in primer binding sites – a feature of the COI marker). In the opinion paper, we don’t dismiss the use of COI, but suggest that binding sites of primers in target groups should be carefully examined and various alternative markers considered. Moving towards multiple markers with some overlapping taxonomic coverage, or using alternate methods of marker recovery, are possible ways forward.

    The figure below shows a visual representation of variation in the primer binding sites in a range of fish species for two potential metabarcoding markers (mtDNA COI on left and mtDNA 16S on right; image taken from supplementary material). Each coloured horizontal line represents a DNA sequence (primer binding site) for a particular fish. Ideally these should be consistent across species, meaning the colours should be lined up in vertical columns:


_____________________________________________

August 4, 2014 - The Molecular Ecology Special Issue: Molecular Detection of Trophic Interactions is now available online. This issue includes a broad range of studies using DNA to study diet. A harbour seal photo from the captive study carried out by Austen Thomas is featured on the cover. There is also a photo of a rock lobster from a study by Kevin Redd who worked with Simon Jarman here at the Australian Antarctic Division.

_____________________________________________

April 18, 2014 - Last week I gave a talk at Deakin University as part of the Centre of Integrative Ecology Seminar Series (Abstract). The focus was on applications of next-generation sequencing in the field of ecology. In particular I talked about DNA metabarcoding for studies of biodiversity (e.g. DNA-based diet studies) and genetic based of adaptive evolution (e.g. genomic divergence across ecological transitions in stickleback fish). 

_____________________________________________

March 18, 2014 -  Kristian Peters' study on Australian sea lion diet was recently published in Marine Ecology (see paper here). The diet of this rare sea lion species is difficult to assess since few prey remains are recovered in their scats (e.g. a previous study found that only 2% of consumed fish otoliths survive digestion). Kristian's paper is the first one to use a genetic approach for Australian sea lion diet analysis. It reveals a broad diversity of prey DNA in the samples, whereas a parallel hard-part analysis of the same scats identified only two prey species!

_____________________________________________

December 20, 2013 -  A study of Adelie penguin diet lead by Simon Jarman was just published in PLOS ONE (see paper here). Scat samples were collected from several Antarctic colonies over a few seasons and were analysed with an assay targeting a conserved DNA marker. A range of expected prey (e.g. krill, fish, copepods, amphipods) as well as a surprising number of other species (e.g. jellyfish and comb jellies) were identified. Differences in diet between breeding stages, colonies and seasons highlight how this non-invasive method will be a useful tool for long-term ecological studies. The figure below taken from the paper shows spatial and temporal comparisons of Adélie penguin population diet.




_____________________________________________

September 23, 2013 -  A paper for the Molecular Ecology Special Issue 'Molecular Detection of Trophic Interactions' with lead author Austen Thomas is now available online. This paper is an extension of our recent work published in Molecular Ecology Resources. It's based on a study of captive seals with a known diet and looks at potential ways to improve quantitative diet estimates derived by sequencing DNA in scats. It primarily focuses on parallel sequencing of prey tissue to help account for technical and biological biases. It also considers using properties of prey (e.g. fat content, protein content, etc) as a proxy for DNA recovery. For example, in this system we found that fish with high fat content are over-represented in the DNA recovered from scats - likely due to reduced digestion of their tissue. This paper does push the boundaries of what is presently being done in this field and understandably it raised the hackles of some reviewers. Such a structured approach  may not be possible when using metabarcoding  to examine highly complex DNA mixtures.

_____________________________________________

July 12, 2013 -  Over the last few months my focus has mainly been on Antarctic krill population genomics, working with a RAD genotyping dataset we produced earlier this year (helped by Floragenex). With over 60 billion base pairs of DNA sequence and 10's of thousands of variable sites, the amount of data is astounding. And we still have more data coming - I just recently finished extracting DNA from krill collected in the Ross Sea (during the Antarctic Blue Whale Voyageand these samples are currently being RAD sequenced. It will be a while before this data will be ready for publication, but I did present some preliminary findings at a very inspiring population genetics course at ANU in Canberra (run by Rod Peakall and Peter Smouse). I also attended a conference on 'Strategic Science in Antarctica' with a broad range of talks on current and future directions of  Australian and New Zealand Antarctic research. 

_____________________________________________

 May 5, 2013 -  A paper I co-authored with UBC PhD student Austen Thomas et al. is now available online (here). It looks at counts of fish DNA sequences recovered from scats of seals fed a know diet, sequencing was done with an Ion Torrent sequencer at the AAD. Our initial intent was to simply confirm that proportions of sequences from fish species could be used as a proxy for composition of a seals' diet (as a verification for field-based studies). However, once we started looking at these data we realised our sequence counts depended on many technical factors (such as direction of sequence read and level of quality filtering). These observations have implications for any study that is using DNA barcoding to analyse DNA mixtures. Anyway, the paper is looking at these technical factors and biases in a bit of detail. All the sequence data and my R code used to analyse it (including code to produce figures in the paper) are available on Dryad (link).

_____________________________________________

April 5, 2013 -  I attended a Biodiversity Genomics Conference at ANU in Canberra, Australia. There were a range of very impressive talks applying next-generation sequencing approaches to study biodiversity, and some very useful discussions in workshops. I gave an invited presentation on our DNA-based diet work, and also presented in a workshop on 'Preparing samples for sequencing and cost-effective library prep'   

_____________________________________________

March 4, 2013 - Molecular Ecology paper on phylogeography and adaptation genetics of Haida Gwaii stickleback populations is now available online (Link). This large survey of genetic variation in >100 populations is a companion paper to an comprehensive analysis of morphological variation: Reimchen TE, Bergstrom C, Nosil P (2013) Natural selection and the adaptive radiation of Haida Gwaii stickleback. Evolutionary Ecology Research 15: 241-269. Tree below shows relationships between stickleback from Haida Gwaii - colours represent different watersheds