The Origin of Species

It is an exciting time to study evolutionary biology and speciation. As a research field that is largely focused on historical events, modern high-throughput sequencing approaches offer unprecedented opportunities to make inferences about said events. This is particularly true for non-model systems with limited experimental options. In combination with theoretical and particularly computational advances, we can now obtain genomes and then attempt to read them “like a history book” for any organism. My research focuses on using genomics to advance our understanding of speciation, and I have been working on this in a variety of non-model organisms: crows, cichlid fish, and lemurs.

The origin of species has always interested me. As a traveling naturalist, it is easy to start wondering about processes behind patterns in geographic distributions and ecology, dissimilarities and similarities between closely related species, and what keeps them apart. Or what evidently doesn't, when you encounter hybrids or clinal variation. I was slow to gain an interest in genetics as an undergrad, and the penny only really dropped during an MSc research project on song divergence as a behavioral mechanism of isolation between two hybridizing subspecies of the Gray-breasted Wood-Wren (see Dingle et al. 2010 and Halfwerk et al. 2016). These birds are morphologically highly similar and vocally divergent, but since their songs are learned, we needed genetic tests to learn the subspecific identity of individual birds. At the same time, this genetic work also revealed high levels of divergence between these subspecies, despite looking nearly identical.

Diversification in Tiny Primates

Mouse lemurs (genus Microcebus) are the smallest primates on earth and are endemic to Madagascar — as is the entire so-called superfamily of lemurs. Until recently, they were thought to comprise only two species, but genetic work has shown that there is much diversity hidden within these highly cryptic, nocturnal creatures. At Duke University, in the lab of my postdoc adviser Anne Yoder and – among others – collaborators George Tiley and Jordi Salmona, I have mostly worked on a large and growing RADseq dataset of all 24 presently described species of mouse lemur to investigate phylogeny, patterns of diversification, demographic histories, and gene flow within and among species.

We have a preprint of a study of five species of mouse lemurs up, which confirms a lineage hitherto called “M. sp. nova #3” as a highly distinct taxon while calling into question the distinctness of two other species. We estimated rates of gene flow (finding, for instance, evidence for mitochondrial introgression) and divergence times. Interestingly, two pairs of these species occur sympatrically without ongoing gene flow, despite having diverged well under a million years ago.

Along similar lines, I have found that previously suspected hybridization between two other mouse lemur species, which also diverged less than 1 ma ago, is not supported by RADseq data — and the lack of hybridization despite co-occurrence implies that these species appear to have rapidly evolved reproductive isolation (unpublished data, with Jörg Ganzhorn). Other ongoing mouse lemur projects include an investigation of genomic divergence associated with local adaptation in reddish-gray mouse lemur (M. griseorufus), modeling divergence and gene flow among all close sister species, and creating a dated phylogeny for all described species. We also have a preprint up on an estimation of the de novo mutation rate for the gray mouse lemur (M. murinus) based on deep whole-genome sequencing of a known pedigree of mouse lemurs from the Duke Lemur Center. Finally, I collaborated with Rachel Williams, a former postdoc in the lab, on a conservation genomic project on dwarf lemurs Williams et al. 2020.

Sympatric Speciation with a Twist

Speciation is often assisted by geographic isolation of populations, after dispersal or vicariance events. Mutation and drift will inevitably lead to genetic divergence when populations are physically separated, while selection will have to overcome gene flow in the absence of geographic isolation. This was generally considered extremely hard if not impossible throughout much of the 20th century. The discovery of several radiations of cichlid fish in isolated Cameroon crater lakes with, within each lake, monophyletic mtDNA (Schliewen, Tautz & Pääbo 1994), marked a real turning point. A flurry of subsequent theoretical work and a search for additional empirical examples ensued, with both lines of inquiry generating additional support that speciation can happen in sympatry. At this point, sympatric speciation is hardly a controversial topic anymore (even though it is generally thought to be uncommon at best). However, we have only recently gained the ability to assess monophyly of sympatric radiations – a telltale sign of sympatric speciation – not just for single genes, but for the entire genome. This is important because signals from individual genes can be highly misleading due to lineage sorting and gene flow.

With all of this in mind, I studied whole-genome sequences of the abovementioned Cameroon crater lake cichlids for a year in Chris Martin's Lab (then at UNC Chapel Hill, North Carolina, now at UC Berkeley, California), working together with PhD student Emilie Richards. This research supported that divergence took place within the tiny crater lakes, i.e. in sympatry, but also suggested that the lakes do not appear to be as isolated as previously thought, given that some secondary gene flow from riverine populations occurred. For one radiation (Poelstra et al. 2018), we additionally found some support for a role of said gene flow in actually promoting speciation, while we did not in another (Richards et al. 2018).

The Curious Case of Crows

Hybrid zones have long been a central locus of speciation research, and are commonly characterized as “natural laboratories of evolution” or “windows into the evolutionary process.” One reason for this is that hybrid zones capture an intermediate stage of speciation, in which reproductive isolation is present but incomplete. A hybrid zone between partially grey (Hooded) and all-black (Carrion) crows in Europe had already been mapped by 1928 (Meise) and was discussed by as a key example of speciation and hybridization by Ernst Mayr mid-century — and subsequently in several modern textbooks on evolution. However, little research had been conducted on this system and had perhaps raised more questions than answers, such as the intriguing finding of an apparent lack of (neutral) genetic differentiation between them.

For my PhD, advised by Jochen Wolf (then at Uppsala, Sweden, now in Munich, Germany), and with much of the work in collaboration with Nagarjun Vijay and Christen Bossu, I looked into this system from a molecular perspective. Initially using candidate coloration genes (Poelstra et al. 2013), and then RNA-seq (Poelstra et al. 2015, and see also Vijay et al. 2013) and whole-genome sequencing (Poelstra et al. 2014). One might argue that once again, more questions were raised than answered, but we generated some interesting findings nevertheless…

For one, Carrion and Hooded Crows are nearly identical across almost the entire genome. This echoes those earlier findings, but is incredibly striking nevertheless, given that phenotypically, they nearly behave as “good species.” Additionally, we found that within Carrion Crows, as well within the phenotypically similar orientalis crows, much higher levels of differentiation can be found. We could also quantify differentiation at a high resolution, finding only 83 fixed differences between Carrion and Hooded Crows them across our entire genome assembly, all but one of which are in a single cluster in the genome. Differentiation in gene expression mirrored this lack of differentiation, with substantial differences only in the bases of feathers from the same body area that differ in color between Carrion and Hooded Crows. Finally, we got a long way towards identifying the genetic changes responsible for these differences in plumage coloration (see also Vijay et al. 2016 and Knief et al. 2019 for increased resolution on that).