Welcome to the web page of the Bank lab at the Gulbenkian Institute. Lab news are displayed below.
Viruses are excellent biological models for understanding how populations evolve because they have characteristics (like high mutation rates and large population sizes) that make it easier to observe evolution in real-time. However, as argued in a new review paper in the journal Heredity, the unusual biology of viruses means that some common, simplifying assumptions of population genetics are not met by viral populations.
In particular, viruses tend to have highly skewed offspring distributions, with some virions producing either many more or many fewer offspring than assumed, and viral populations often experience drastic changes in size (i.e. bottlenecks) as a result of transmission or infection. These features of virus biology mean that the Kingman coalescent and Wright-Fisher model (part a in the Figure below) that are traditionally used in population genetics can lead to an erroneous inference of how the virus populations are evolving. The authors argue that the multiple merger coalescent class of models (part b of the Figure) can account for these limitations of traditional models by allowing more than two lineages to coalesce at a time.
New computational approaches, such as the use of multiple merger coalescent models or forward simulations, will elucidate how the unusual biology of viruses influences their genomic diversity and evolution.
Figure: Each row of dots shows the alleles in a single generation, with the lines connecting dots showing reproduction events. For each type of coalescent (a and b), the left panel shows the evolutionary process of the whole population, whereas the right panel shows a possible sampling and its genealogy. Unlike the Kingman coalescent (a), the multiple merger coalescent (b) allows parents to give rise to more than two offspring in the next generation. (Figure taken from figure 1 of the paper.)
–This summary was written by Telmo Cunha & Hermina Ghenu
One interest of our lab is predicting where evolution will go, in the genotype space, and which paths evolution can take to improve fitness. To this end, fitness landscapes are a useful concept for understanding the predictability of evolution since they relate the genotype to the reproductive success of an individual. Empirical fitness landscapes are especially important because they allow us to compare real data with the theoretical work that has accumulated over more than 80 years of research.
The latest collaboration between the Bank and Jensen labs examines an unprecedentedly large empirical fitness landscape composed of amino-acid changing mutations in the heat shock protein Hsp90 in yeast. Our collaborator Ryan Hietpas generated and screened mutants that are up to six amino-acids away from the parental genotype, allowing the authors to look at the interactions between mutations (epistasis) and compare subsets of the data (local landscapes) to the entire, global landscape. Their main findings are:
- The landscape is dominated by epistasis. Most of the landscape experiences antagonistic epistasis, where mutations that are neutral or beneficial when alone have a detrimental effect when they are found together. In contrast, the global fitness peak of the landscape is the result of synergistic epistasis, where mutations that are neutral or slightly beneficial when alone yield high reproductive success in combination.
- Studying potential adaptive walks on the landscape shows that although the global peak can theoretically be reached most of the time, adaptation may stall at an intermediate fitness peak, which is reached with high probability from the parental genotype.
- Analyses of the entire set of mutations by comparing estimated landscape statistics to expectations from theoretical landscape models suggest that the landscape is heterogeneous and its topography is globally hard to predict.
This work highlights the dual nature of epistasis: a landscape with no interactions between mutations is just a single global peak (globally predictable), and evolution will always end up at the top but it is impossible to know what paths it will take (locally unpredictable). Conversely, a landscape dominated by interactions has few viable paths (locally predictable) but each one leads to a different peak (globally unpredictable).
The empirical fitness landscape studied here suggests that our current theoretical understanding of fitness landscapes is not sufficient to describe empirical landscapes. More generally, this work draws attention to the importance of comparing and informing theoretical models with empirical data, and proposes approaches to improve this process in the future. The paper is available online in PNAS or as a preprint.
The last couple of weeks have been so busy that I am only now catching up with posting our lab news, including welcoming our two new lab members, Mark and Alexandre. Mark is a biochemist who will be visiting the lab during the next few months for an exchange about the evolutionary versus biochemical implications of empirical fitness landscapes. Alexandre has joined us as a postdoc; he is an evolutionary modeler who is particularly interested in the role of Dobzhansky-Muller incompatibilities in speciation and hybridization.
In this new paper in Evolution the authors study the evolutionary dynamics of influenza A virus under different concentrations of Favipiravir, which is a drug that leads to an increase in mutation rate across the genome. By tracking down real-time evolution of several populations they are able to evaluate the extinction dynamics and the potential adaptive response of the virus to different drug treatments.
With this setup the authors were able to show that:
1) Extinction occurs under high mutation rate – The virus populations under an increasing drug concentration show an increased mutation rate and number of mutations accumulated, resulting in rapid extinction and providing support for mutational meltdown as driving mechanism;
2) The virus populations may be able to adapt to intermediate drug conditions – Populations subjected to constant intermediate concentrations of Favipiravir showed indications of an adaptive viral response, suggesting that resistance may emerge under specific drug treatments;
3) Evolution of drug resistance can be explored by a combination of population genetic models and experimental evolution – The combination of population genetic models and experimental evolution is an excellent means to understand the evolutionary dynamics and genetics of virus resistance to drug treatments and to test the efficacy of new possible treatments.
A lot will be going on in the lab in September and October, and this only the beginning:
Claudia will be spreading the word about some of the beautiful things we are doing during next week’s The Ecology of Genome Evolution Symposium at Uppsala University in Sweden.
Inês is leaving us tomorrow and will spend the whole month working in Dan Bolon’s laboratory at the University of Massachusetts Medical School.
And last but not least, Hermina is presenting a poster about hybridization dynamics in ants on EMBO Young Scientists’ Forum, which is happening today and tomorrow at the Fundação Calouste Gulbenkian in Lisbon.
Registration is now open for a training course in evolutionary modeling in the framework of the GTPB, which will take place at the IGC in November (make sure to allow some extra days for sightseeing and surfing). Below is a description of the course, or read the extended version (and apply quickly – there are only 20 spots!) here.
Applied Evolutionary Theory
A hands-on introduction to creating and analyzing models of evolution
with Claudia Bank (IGC), Rafael Guerrero (Indiana University) and Stephan Peischl (University of Bern)
For many of its history, our knowledge of evolution has been based heavily on theoretical models and hypotheses. In the age of novel experimental and technological approaches, we are now increasingly able to evaluate this theory; however, the basics of how and why to develop and analyse a simple model are often forgotten in the process of NGS analysis. This course aims at training evolutionary biologists in classical modelling and teach them ways to approach their own research questions through evolutionary theory.
Primarily through interactive hands-on sessions, complemented by an introduction to the cornerstones of modelling and its application to data analysis, this course will familiarize the participants with ways of approaching a research question with a simple model, and different strategies at gaining insight from the model. In groups of two, course participants will develop and analyse their own toy model in the course and present their findings to the group on the last day.
Topics that will be covered in the course include the following:
Why and how are models useful?
How to write down/develop a model
How simple/complicated should a model be?
Which modelling approach/programming language should I use for my question?
How to nail down a question with a model
Extracting results from an equation/simulation
How to evaluate a model using empirical data
Participants can use their preferred programming language during the hands-on sessions, and free access to Wolfram Mathematica will be provided. The instructors have modelling experience using Mathematica, R, Python, and C++.