Last semester's seminars
30-May-2013 -- Tami Lieberman (HMS, Kishony Lab) - Bacterial diversification within individual patients: Recent studies tracking bacterial evolution within individual patients have identified mutations responsible for antibiotic resistance, the rate at which mutations accumulate, and specific bacterial genes and pathways that are under selective pressure. Such studies typically use sequencing of a single colony isolate from each clinical sample, giving only a limited picture of the diversity of the pathogen population at the time of sampling. In this talk, I will describe our latest results, using colony resequencing and deep population sequencing of Burkholderia dolosa from people with cystic fibrosis, describing the intrastrain genomic diversity within individual clinical samples.
16-May-2013 -- Kevin Esvelt (Wyss Institute) - Controlling fitness landscape topology in real-time: Evolutionary discovery is constrained by the relative accessibility of nearby fitness peaks, precluding the discovery of distant solutions isolated by fitness valleys. Using phage-assisted continuous evolution as an accelerated evolution platform, we are exploring methods of reshaping the landscape to favor exploration away from already-discovered solutions. By automatically monitoring evolved activity levels to identify when a fitness peak has been located, we seek to promote escape by inducing genetic drift or even inverting the landscape while blocking specific undesired paths via RNA-guided nucleases.
2-May-2013 -- Nick Leiby (Harvard, Marx Lab) Metabolic Tradeoffs and Specialization in Escherichia coli Long-Term Evolution: Evolutionary adaptation to a new environment is often accompanied by specialization and tradeoffs, whereby there is reduced fitness in other environments. We studied strains of E. coli evolved in parallel for 50,000 generations in a stable glucose environment, a scenario where specialization and tradeoffs are expected. However, direct measurements of growth rate contradict this expectation. We found that tradeoffs were neither as widespread nor as parallel as previously reported, and surprisingly, there are approximately as many performance increases as decreases on non-glucose substrates. I will discuss these data and their implications, as well as my ongoing project designed to specifically address the evolutionary canalization of metabolic function.
18-April-2013 -- Julien Ayroles (Harvard) - Using Experimental Evolution to Study the Temporal Dynamics of Allele Frequency Change Over Time: For nearly a century, experimental evolution has been a favorite tool of biologists seeking to test various evolutionary models. This approach can provide an unprecedented understanding of the dynamics of evolutionary change. What could we learn if one could track changes over the entire genome? We sequenced DNA pools from lines of flies selected for starvation resistance and compared them to control lines every generation. After only 15 generations of selection, we observed a strong phenotypic response associated with changes in allele frequencies in the selected lines. I will discuss how such data can be used to describe the temporal dynamics of allele frequency change. I will also contrast the complex genetic architecture of starvation resistance in the unselected population to the surprisingly simple genetic architecture associated with lines selected for increase starvation resistance.
11-April-2013 -- John Barton (MIT, Chakraborty Lab) - From HIV sequence data to fitness landscapes: Development of an effective vaccine against HIV, the greatest hope for controlling the epidemic, has been hindered by the extreme mutability of the virus. To identify regions of vulnerability in the viral proteome, we have constructed a computational model to infer fitness landscapes for HIV proteins from publicly available sequence data. I will discuss experimental tests of these proposed fitness landscapes, and early applications of this technique in understanding intra-host viral evolution.
28-March-2013 -- Anna Selmecki (HMS) - How polyploidy and aneuploidy impact the speed of adaptation: Polyploidy, having more than a diploid set of chromosomes, is frequently found in nature and has played an important role in the evolution of new species. Theoretical work indicates that ploidy will affect adaptation, and experimental work in haploids and diploids has supported much of this theory, but little is known about the rate and molecular basis of polyploid adaptation. We have compared the evolutionary trajectories of isogenic haploid, diploid, and tetraploid yeast and found that tetraploids adapted more rapidly than lower ploidy cells. Our whole genome analyses indicate that the adaptation was driven by an increase in acquired mutations, polymorphisms and chromosome level changes.
14-March-2013 -- Daniel Rice (Harvard, Desai Lab) Sequence-level dynamics of adapting yeast populations: The dynamics of adaptation determines which mutations fix in a population, and hence how reproducible evolution will be. High time resolution sequence data can shed light on these dynamics. I will present whole-genome whole-population sequence data of 40 replicate laboratory yeast populations. I will discuss how patterns of sequence evolution are driven by a balance between chance hitchhiking and interference on one hand, which increase stochastic variation in evolutionary outcomes, and the deterministic action of selection on individual mutations, which favors parallel evolutionary solutions in replicate populations.
Older seminars (2010-2012)
29-Nov-2012 Manoshi Datta (Gore Lab (MIT)): Range expansions favor the evolution of cooperation. Natural populations throughout the tree of life undergo range expansions in response to changes in the environment. Recent theoretical work suggests that range expansions can have a strong effect on evolution, even leading to the fixation of deleterious alleles that would normally be outcompeted in the absence of migration. However, little is known about how range expansions might influence alleles under frequency- or density-dependent selection. Moreover, there is very little experimental evidence to complement existing theory, since migrating populations are extremely difficult to study in the natural environment. In this study, we have used a yeast experimental system to study the effect of range expansions on the evolution of cooperative behaviors, which are widespread in nature and commonly display both negative-frequency dependence and density-dependent growth. We found that range expansions favor the evolution of cooperation in two ways: (1) through the enrichment of cooperation at the front of the expanding population wave, and (2) by allowing cooperators to “outrun” an invading wave of defectors in a spatially extended population. We conclude that cooperation is enhanced through the coupling of population ecology and evolutionary dynamics, thus providing experimental evidence for a novel mechanism through which cooperative behaviors could be maintained in nature.
15-Nov-2012 Justin Meyer (Kishony Lab): Evolution before and after a key innovation in phage lambda. The evolution of novelty has long captured the imaginations of biologists for two reasons: 1) these events are hard to explain with population genetic theory, and 2) moments of innovation are thought to spark rapid evolutionary change including speciation. Recently, colleagues and I have developed a model microbial system to study the evolution of novelty in the virus, bacteriophage λ. For this talk, I will review results published earlier this year on the evolutionary processes that led to an innovation in λ, as well as new findings on this system. Previously, we reported that λ co-cultured with its host Escherichia coli repeatedly evolved to exploit a new protein receptor, OmpF. Mutations necessary for this novel function were shown to fix by the action of natural selection and whether they occurred depended on the coevolutionary path the host took. For this talk, I will provide a more quantitative picture of how the phage evolves, including measurements of λ’s fitness landscape and how the landscape is distorted during the process of coevolution. These details will help pinpoint where theory on novelty often falls short. Additionally, I will explore what happens after the innovation evolves, including a coevolutionary arms race that drives rapid molecular evolution, and eventually sympatric speciation.
8-Nov-2012 Edel Hyland (Murray Lab): The Molecular Evolution of Plasmids. Plasmids are categorized into groups, or species, based on their compatibility with other plasmids; those that can co-exist within the same cell are said to be compatible and represent two unique plasmid species, whereas those that are in competition with each other, are incompatible and are thus the same species. This competition is dictated in part by the plasmid's segregation machinery. Our goal is to understand the molecular basis of plasmid compatibility, and to ask how many interfaces in the interactions between components of the segregation machinery are evolutionarily malleable, in order to facilitate the differentiation into new plasmid species.
18-Oct-2012 Bryan Dickinson (Liu Lab): Divergent selection and stochasticity limit subsequent evolutionary convergence in experimental protein evolution. Evolutionary outcomes are unpredictable if adaptation is sensitive to either transient differences in the selective environment or to the stochastic appearance of mutations. Here we apply a new model system to investigate evolutionary contingency and convergence at the molecular level. We combined phage-assisted continuous evolution (PACE) with high-throughput sequencing to determine whether identical populations of a single protein subjected to divergent selection pressures will exhibit parallel genotypic and phenotypic evolution when subsequently selected for convergence. Independent populations of phage-encoded T7 RNA polymerase were evolved over hundreds of generations to first recognize either the T3 or the SP6 promoter, then to recognize a hybrid TP6 promoter. We observed distinct classes of genetic solutions with unequal phenotypic activity and evolutionary potential arising from the two selection regimens, as well as differences among replicate populations exposed to the same regimen. Mutational analysis identified the causative epistatic interactions precluding genotypic and phenotypic parallelism. Our results suggest that transient adaptation to different environments generates unpredictable molecular diversity even after convergent evolution.
4-Oct-2012 Jan-Hendrik Hehemann (Polz Lab, MIT): Evolution of glycan metabolism in human gut bacteria. Humans host an intestinal population of microbes – collectively referred to as the gut microbiome – which encode the carbohydrate active enzymes, or CAZymes, that are absent from the human genome. These CAZymes help to extract energy from recalcitrant food polysaccharides. The question then arises if and how the microbiome adapts to new carbohydrate sources when modern humans change eating habits. Recent metagenome analysis of microbiomes from healthy American, Japanese and Spanish populations identified putative CAZymes obtained by horizontal gene transfer (HGT) from marine bacteria which suggested that human gut bacteria evolved to degrade algal carbohydrates for example consumed in form of Sushi. We found these CAZymes are transferred in form of large genetic clusters and we study such mobile genetic elements using biochemical, structural and molecular biology approaches to understand how they transform the abilities of recipient microbes and help us to digest our foods.
20-Sep-2012 Adam Palmer (Kishony Lab): High-order genetic interactions between drug resistance mutations produce a multi-peaked adaptive landscape. To understand how microbial adaptation to antibiotics can result in equally resistant phenotypes from diverse genotypic changes, we synthesized and phenotyped all combinatorial sets of many trimethoprim resistance mutations. We found that the adaptive landscape is characterized by high-order genetic interactions where the pairwise compatibility/incompatibility of mutations depended upon the presence of yet other mutations. The evolution of drug resistance thus avoids the theory that 'reciprocal sign epistasis' is necessary for multiple adaptive peaks, as these high-order interactions yield multiple peaks despite a lack of any consistent reciprocal sign epistasis.
May 31st 2012 Kirill Korolev (MIT) Deleterious mutations in cancer
Abstract: Cancer is an outcome of somatic evolution. To outcompete their benign sisters, cancer cells need to acquire many heritable changes (driver mutations) that enable proliferation. In addition to the rare beneficial drivers, cancer cells must also acquire neutral or slightly deleterious passenger mutations. Indeed, the number of possible passengers exceeds the number of possible drivers by orders of magnitude. Upon including passengers in our model, we found that cancer is no longer a straightforward progression to malignancy. In particular, there is a critical population size such that smaller populations accumulate passengers and decline, while larger populations accumulate drivers and grow. The transition to cancer for small initial populations is, therefore, stochastic in nature and is similar to diffusion over an energy barrier in chemical kinetics. We also found that there is an optimal mutation rate for cancer development, and passengers with intermediate fitness costs are most detrimental to cancer. Our theory can explain some paradoxes in cancer research and is consistent with recent sequencing data. Finally, we show that cancer's position in the parameter space can be crucial for selecting the most successful treatment.
May 17th 2012 Ariel Weinberger (Gilmore lab) Adaptive Immunity: Why some microbes have it and others don't.
Bacteria and archaea have elegant adaptive immune systems that target and destroy specific viral and plasmid DNA sequences. Termed CRISPR, these adaptive immune systems are found in over 90% of archaea but only 50% of bacteria. I'll combine genomics and modeling to try to explain the relative lack of CRISPR in bacteria, capturing thresholds in the viral mutation rate above which CRISPR cannot evolve.
May 10th 2012 Genya Frenkel (Desai lab): Tracking dynamics of adaptation reveals the distribution of fitness effects and evolution of ecology in asexual budding yeast
Tracking the frequencies of neutral markers in a population is a classic way to observe the dynamics of adaptation, in particular the appearance of beneficial mutations and their subsequent trajectories toward fixation or loss. I will present the results of a relatively large-scale study of this kind. Into ~1000 identical populations of budding-yeast, I seeded mutants with known initial fitness advantages and frequencies. I observed their trajectories to be at-first deterministic, as expected, and subsequently highly variable. The goal is to infer from the statistics of these outcomes (e.g. fixation probabilities) information about the rate and distribution of fitness effects of beneficial mutations. In some cases, the populations evolved into two types that co-exist at a stable equilibrium ratio, which was found to be dependent on the spatial structure of the population at the bottom of microtiter wells.
April 19th 2012: Wolfram Moebius (Nelson lab): Range expansions into heterogeneous environments
Wolfram says: "I'm interested in how a species spreads into new territory which is not homogeneous, but heterogeneous (think as mountains or lakes as obstacles in the plains). To investigate this question, I try to use a combined theoretical and experimental approach using a model system of bacteriophage T7 spreading on a (heterogeneous) lawn of E. coli."
April 5th 2012: John Coffin (Tufts University): Host-retroviral coevolution
John Coffin is a professor at Tufts and the director of the HIV Drug Resistance Program of the National Cancer Institute, going back and forth between Boston and Bethesda almost every week. He is also a member of the National Academy of Sciences. John is well knwon for his work on HIV. His 1995 Science paper is hugely influential (with >1600 citations) and many of his newer papers on HIV are important too. However, he has also done a lot of work on other retroviruses including endogenous retroviruses. Such viruses are present in high numbers in our genome and can be used to study our evolutionary history as well as the history of the viruses. This is what he will talk about.
March 22 2012 Liedewij Laan (Murray lab): How evolvable are polarization mechanisms?
ABSTRACT: Why do cells simultaneously use different polarization mechanisms and why do even closely related species seem to use a different variety of them? We removed genes essential to a specific polarization mechanism in yeast and subsequently evolved these “nearly dead” cells to significantly higher fitness. I will present our first attempts to study how these cells fixed the “deleted” mechanism.
March 1 2012 Alison Hill (Nowak lab): Predicting and preventing drug resistant HIV at multiple levels of selection
ABSTRACT: The evolution of drug resistance in HIV is an example of selection occurring at multiple spatial and temporal scales. First, I will discuss techniques we have developed to study the emergence of drug resistance within a host, including adapting the mutant selection window to virology and attempts to realistically simulate clinical trials. I will present results on how pharmacodynamics and pharmacokinetics of antiretroviral drugs affect the generation and selection of resistance mutations, and how this is influenced by viral latency, drug-protected compartments in the body, and different modes of cell-to-cell transmission. Secondly, I will discuss the population wide evolution of HIV, including the potential for a drug-resistant epidemic and models for how the spread of new disease variants in influenced by population structure.
February 23 2012 Gregg Wildenberg (Murray lab): Evolution of a Circadian Clock in Yeast
ABSTRACT: I seek to understand the molecular basis of evolutionary novelty. I have experimentally evolved a circadian clock in Saccharomyces cerevisiaeto ask how an organism evolves a new function and will identify the molecular and cellular changes that gave rise to this novel behavior.
February 9 2012 Adrian W.R. Serohijos (Shakhnovich lab): Imprints of protein biophysics on the molecular clock
ABSTRACT: Evolution is a dynamic process acting at multiple length and time scales—single mutations affect the molecular properties of DNA/RNA/proteins, consequently increasing or decreasing organismal fitness. The mutation’s fate is likewise dictated by population size since species survive or become extinct in the wild as a population. I will describe our efforts to create a systematic framework that bridges these scales. I will also present some results that address outstanding questions on the nature of the molecular clock.
January 26th 2012 Sergey Kryazhimskiy (Desai Lab): Champions and losers in a race to the top: epistasis and adaptation in yeast.
ABSTRACT: We know that populations adapt when they encounter new environments, but our ability to predict the course and rate of adaptation are currently very limited. One complication is that we do not know how one mutation might influence the fitness effects of other mutations. In other words, we do not know the structure of epistasis among beneficial mutations. I will describe some experiments that are aimed at probing the statistical structure of epistasis among beneficial mutations and the implications of our findings for the predictability of adaptation.
December 12 2011 Tim Van Opijnen (Tufts University): The virulence landscape of a bacterial pathogen. Tim's Website.
November 28 2011 Nigel Delaney (Marx lab): Insights into evolution from the fastest evolving bacteria on earth.
November 14 2011 Will Harcombe (Marx lab): Evolution of complex networks: Testing the predictions of flux balance analysis.
October 31 2011 Melanie Müller (Nelson lab): Range expansion of mutualists.
October 3rd 2011 Hsiao-Han Chang (Hartl lab): Population genetic inferences of Plasmodium falciparum on fully sequenced genomes from Senegal.
September 19th 2011 John Koschwanez (Murray lab): Sticking together makes life sweeter. See recent story in Harvard Gazette.
June 13th 2011 Rolf Ypma (The Netherlands Institute of Health): "A statistical framework for reconstructing transmission trees: avian influenza as a case study"
June 1st 2011 Daniel Rosenbloom (Harvard, FAS, OEB) "The Good, the Bad, and the Ugly: Predicting HIV treatment success, failure, and drug resistance using a model of virus dynamics"
March 21st Tami Lieberman (Harvard, HMS, Systems): "Genomic evolution of a bacterial pathogen within human hosts"
February 22nd 2011 Gabriel Perron (Harvard, FAS, Systems): "The evolution of antibiotic resistance: insights from field and experimental studies"
February 7th 2011 Scott Wylie (Harvard, FAS, Chemistry): "How molecular biophysics constrains fitness landscapes"
December 13th 2010 Pleuni Pennings (Harvard, FAS, OEB): "Modeling the evolution of drug resistance in HIV"
November 30th 2010 Deepa Agashe (Harvard, FAS, OEB): "Selection against synonymous mutations"