Posted on 11th August, 2018 by Carlo Pecoraro
Speciation is a fundamental process responsible for creating the diversity of life on the Earth. How do new species form? Darwin was the first one who tried to explain this process, arguing that
natural selection was the driving force of speciation, intrinsically augmented by ecological conditions (The Origin of Species (1859)). While the introduction of Mayr’s (1942) Biological Species
Concept, however, underlined the difficulties of establishing reproductive isolation without prolonged geographical separation.
These two theories, sympatric vs. allopatric, are the two main views of speciation but they were considered to be opposed for a while. Nowadays, it is appreciated that they are both fundamental driving forces of speciation and genomic approaches are serving to enhance understanding of the interacting extrinsic and intrinsic forces that drive speciation.
Here we have the possibility to discuss about this fascinating topic with our instructors Dr. Joana Meier (University of Cambridge, UK) and Dr. Mark Ravinet (CEES, University of Oslo, Norway) of the “Speciation Genomics” Workshop that we will run this December (03-07 December 2018) in Berlin!
When and where did you start working on speciation genomics?
JOANA: I started working on speciation genomics during my Master project with Gerald Heckel and Laurent Excoffier (population genetics group, University of Bern, Switzerland). I sequenced AFLP markers to study patterns of introgression in a hybrid zone of the common vole. During my PhD and first postdoc with Ole Seehausen, I studied speciation in the fastest adaptive radiation known, that of Lake Victoria cichlids (fish ecology and evolution group, EAWAG and University of Bern). I started with RAD sequencing and quickly moved to whole genomes. Currently, I work with 450 genomes of cichlids from Lake Victoria and surroundings. In October, I will move to Cambridge to work on speciation genomics in Heliconius butterflies. The study systems kept changing, but the key questions remained the same throughout my research career: What genomic properties facilitate speciation and why do some lineages speciate much more frequently and quickly than their close relatives?
MARK: I became interested in speciation as a PhD student back in 2008. I was originally hired to do a project on trophic ecology of sticklebacks but found speciation a much more interesting topic! I started working on population genetics at the same time and just as my PhD ended, high throughput sequencing was taking off in the field, so as a postdoc, I quickly switched to working on genomic data… and I haven’t looked back since!
In your opinion, which are the most influential works in speciation genomics so far?
JOANA: Classical work by Ernst Mayr, Bateson-Dobzhansky-Muller, Chung Wu, Nick Barton, Jerry Coyne, Allen Orr, Sergey Gavrilets, R+P Grant, D+B Charlesworth and others have set the stage for speciation genomics. Since the advent of high-throughput sequencing technologies, empirical studies of speciation genomics took off and raised new questions about the genomics of speciation. Papers by Patrick Nosil, Jeff Feder, Sam Yaeman, Chris Jiggins, Ole Seehausen and others are shaping our expectations about genomic signatures of speciation.
MARK: Hmm, this is a much tougher question than I thought it would be. I think there are two categories here – one is a set of theoretical papers which really predate the advent of high throughput sequencing. So, a lot of the work by Nick Barton, Joe Felsenstein, Allen Orr and the Charlesworths for example laid the ground for what we might expect when look at genomic data in the context of speciation. The other category is papers that have really exploited a combination of a cool study system and new technologies. Marius Roesti’s papers on lake-stream stickleback divergence are some of my favourite. I also think there is some excellent experimental work combined with genomics by people like Patrik Nosil and Scott Egan. Also some of the recent work on flycatchers by Reto Burri, Hans Ellegren and colleagues has been really important for pointing out some issues we need to address.
Can you briefly describe us your current project/s on
JOANA: Currently, I am working on cichlids from the Lake Victoria region. I am particularly interested in the role of hybridisation in the rapid evolution of this amazing species diversity. During my PhD, I discovered that all Lake Victoria region cichlids are derived from two ancestral lineages that hybridized at the onset of this massive radiation. Currently, I am studying how hybridization facilitated the adaptive radiation in Lake Victoria after its refilling just 15,000 years ago. The 500 species that evolved in this short time span seem to be new assemblies of the old genetic variation. From 1st October, I will also work on the role of introgression in adaptation and speciation in Heliconius butterflies. Methods-wise I focus on population genomics analyses, demographic modelling, phylogenetics, tests of hybridisation and genome scans.
MARK: I am currently working on the role of gene flow and hybridisation in speciation. I still work on the species that started me off in the field, sticklebacks, but I am working more and more on sparrows. I am also starting to look more at how human activity can shape evolution – particularly the evolution of human commensalism in house sparrows. I think this has the potential to give some really interesting insight into the speciation process.
How do you think NGS has changed the way we study the ecological and evolutionary forces that drives a speciation process?
JOANA: High-throughput sequencing technologies have allowed us to complement candidate gene approaches with studies of genome-wide patterns of adaptation and speciation even in non-model organisms. This allows us to generalize previous findings from single gene studies in model organisms. We learned that the genetic basis of speciation is generally very complex and often involves old genetic variation derived from hybridisation. Originally, hybridisation between species was thought to be extremely rare. However, recent studies show more and more that hybridisation is widespread and often contributes the raw material for speciation in other species. It is now also possible to study the genomic architecture of speciation, i.e. how many genes underlie divergent adaptation or reproductive isolation and how they are arranged on the chromosomes. In addition, we can now more easily identify gene duplications, rearrangements, recombination rate evolution, and transposons. We are only beginning to understand how such genome properties affect adaptation and speciation.
MARK: Yes absolutely. We have traditionally viewed gene flow as something really detrimental to the speciation process. However, we are starting to view gene flow as much more of a creative force. Hybridisation can generate new novelty for strong selection to act upon for instance. I also think we are learning that many of the genes and alleles involved in speciation are actually much older than the species themselves. I think we are starting to move away from simplistic species concepts and towards a better understanding of just how complicated and multi-faceted the process really is.
The advent and improvement of long-read sequencing by Third Generation Sequencing methods such as PacBio and Nanopore have shown promise in producing high-quality assemblies for complex genomes– In your opinion, how will long-reads sequencing technologies affect speciation genomic studies?
JOANA: Most early speciation work was limited to Drosophila flies and other model organisms for which genomic tools were available. With the advent of long-read sequencing technologies it is now feasible to produce high-quality assemblies of non-model organisms. Once costs of long-read sequencing technologies decrease, many animal and plant systems will become accessible for speciation genomic studies. This will allow us to generalize findings from the main study systems and thus provide a better understanding of the origin of species.
MARK: The most immediate way is much better and easier genome assemblies. The easier it is to have a good level assembly, the greater your power for analysis. It remains to be seen how these technologies could be leveraged for population genomics but I think there is potential for improving things like phasing too, which also provides another level of detailed inference. I am looking forward to the day these technologies are so accessible, all our samples can be long read…
Thanks guys for your time. See you in Berlin!!