Projects and Scientists

Biological Diversity: Generation, Control and Exploitation

We will study the generation, control, and exploitation of diversity at multiple scales, ranging from the effects of alterations in the structure of single molecules, through behaviors produced by variations in biological networks (functional modules), to the consequences of diversity within populations of organisms. Several of our research groups will span one or more of these scales and there are collaborations between groups at every level of organization, but for convenience, we divide the proposal into sections for each level of biological organization: molecules, modules, and organisms.

Molecular Diversity

Three projects deal directly with variation in the structure of single molecules.

Andrew Murray, the Center’s principal investigator, and David Nelson will use two complementary approaches to ask how mutations alter reproductive fitness. Reconciling evolutionary theory with field studies and experimental evolution requires that we know the probability of getting beneficial and deleterious mutations of different strengths and understand how pairs of mutations interact with each other and how the costs, benefits, and interactions between mutations changes with the environment. This information is now available for the set of gene deletions in yeast, but not for the random mutations that are likely to dominate evolutionary change.

Kevin Verstrepen will ask how the expansion and contraction of tandemly repeated sequences produces diversity. Tandem repeats are noisy genetic elements that produce phenotypes that vary more slowly than epigenetic variation but faster than classical mutations. They allow a population of individuals to express a diversity of phenotypes, and selection on this diversity can rapidly alter the mean properties of populations.

Allan Drummond will study how the sequences of proteins are dictated by the requirement that they fold as often as possible into a functional molecule and ask how these constraints explain the slower evolution of highly expressed proteins. Our hypothesis is that evolutionary rates of proteins are constrained by translation errors and misfolding rates, and that highly expressed proteins must fold robustly so that translation errors do not produce toxic aggregates.

Network diversity

We will study the diversity and evolution of genetic networks at four levels, theoretical analysis of the limits of biological control circuits, the variation in gene expression from a single promoter, the behavior of signaling networks that control reproduction and survival, and the construction of artificial networks to report on internal cellular states.

Johan Paulsson will ask how the combination of noise and time delays limit the precision of biological regulation. He will use a simple, but powerful, approach to define these theoretical limits for different control strategies.

Michael Elowitz and Naama Barkai will evolve promoters. Elowitz will test the theoretical prediction that very strong selection for increased gene expression will select for increased noise in gene expression. Barkai will test the empirical suggestion that promoters containing TATA elements show noisier gene expression and are thus easier to evolve to increased mean expression levels.

Sharad Ramanathan will use meiosis in budding yeast to investigate the variability in the timing of developmental processes and ask whether such variability can optimize the trade-off between reproduction and survival in unpredictable environments.

Michael Brenner and Katharina Ribbeck will examine how diverse surface properties of proteins affect their ability to enter and exit the nucleus through nuclear pore complexes. The hypothesis is that surface properties encode the efficiency of nucleocytoplasmic transport, which limits the rate at which information can be communicated from the environment to the genome. Thus, diversity of protein surface properties may directly regulate nuclear responses such as gene expression.

Andrew Murray and Naama Barkai will investigate and evolve the mating pathway of budding yeast. Yeast cells mate efficiently under a wide range of conditions, and the evolution of this pathway contributes to speciation. Murray and Barkai investigate the biology and evolution of mating focusing on how cells can show robust responses despite noisy gene expression, how cells mate robustly in noisy environments, and how selective pressure can alter the properties and functions of a network that has been evolved to show these forms of robustness.

Tim Mitchison will study the response of cancer cells to anti-mitotic drugs. Cancer arises, progresses, and evades therapy by accumulating mutations that divert cells from their normal behavior. Mitchison will investigate the physiological and genetic differences that explain why some cell lines die and others survive when treated with spindle poisons.

Yaakov Benenson constructs logical circuits inside cells. The ultimate demonstration that we understand processes is that we can reconstruct them. Benenson embeds artificial logical circuits in cells that convert cellular inputs into fluorescent outputs. He investigates stability and diversity in these circuits and will develop methods to stabilize their outputs to variation in the stoichiometry of their components, with the goal of producing an artificial Boolean evaluator to read out complex patterns of miRNAs and transcription factors in stem cells and differentiated cells.

Organismal diversity

We will study how populations of organisms change in response to natural selection in the field and the lab and study the evolutionary changes that affect host-pathogen interactions.

Marcus Kronforst will identify genetic differences that cause speciation. He studies Heliconius butterflies, which provide a powerful tool to study the earliest stages of reproductive isolation. Closely related species can mate with each other making it possible to find the genes responsible for the patterning differences and different mate choices that have caused speciation.

Katharina Ribbeck and Kevin Foster will study the interaction between Pseudomonas aeruginosa (an important pathogen of cystic fibrosis patients) and mucus. Ribbeck studies mucus as a dynamic structure that controls the transport of molecules and organisms between the extracellular fluid and the surface of animal cells, and Foster studies the social biology and evolution of microbes. They will collaborate to examine the social behavior of P.aeruginosa on mucus and compare the evolution of this organism on mucus in the laboratory and in the lungs of patients.

Irene Chen will ask what limits the host range of pathogens. The host range of filamentous phage is determined by a single tail protein, which she will mutagenize to evolve the host range of the phage with the aim of determining the influences of selection and genetic drift in restricting host range in natural populations.

Roy Kishony will study the ecology and spread of antibiotic resistance in soil and the behavior of artificially constructed microbial communities. Random collections of soil bacteria exhibit a wide range of resistances to antibiotics and serve as a genetic reservoir that could seed pathogenic bacteria via horizontal gene transfer. Despite their importance, little is known about the spatial structure, species-species interactions, and responses to drug-induced perturbation in these communities, which serve as useful models of human microflora.

Former Projects and Scientists

Aviv Regev, in collaboration with Nir Friedman and George Church, used computational approaches to find modules, analyze their structure and function, and reconstruct their evolution. Recent accomplishments include:

Module Map, a computational approach for associating the regulation of modules with biological conditions; for cancer, this approach identified collections of molecules that shed new light on metastasis (Segal et al. 2004; Segal et al. 2005);
showing that, although the genes in transcriptional modules are evolutionarily conserved, the transcription factors and DNA motifs that control their expression are not (Tanay et al. 2005);
GeneXPress, a framework for analyzing modules that allows users to find modules, see their components, and compare the effects of different definitions of modules (Segal et al. 2004).
SYNERGY, and algorithm that reliably distinguishes orthologs from paralogs on a genomic scale; its use led to the identification of seven principles that govern copy number variation across 300 million years of fungal evolution, and clarified the contribution of copy number variation to genome and module evolution (Wapinski et al. 2007).

Daniel Fisher built theoretical models, and using them to predict and analyze the behavior of modules in vivo. This work has led to improved understanding of the generation of segment polarity in Drosophila (I ngolia 2004) and, in collaboration with the Mitchison lab at Harvard Medical School, to a model for meiotic spindle organization in Xenopus egg extracts (Burbank et al. 2006, 2007).

Kurt Thorn has developed methods to monitor protein modification and interactions in yeast, using fluorescence resonance energy transfer (FRET) microscopy (Sheff & Thorn 2004). He has also used these techniques in a collaboration with Michael Laub, to find interactions between two-component signaling proteins in the bacterium Caulobacter crescentus.

Michael Laub used genomics, genetics, biochemistry, cell biology, and computational biology to study the structure and function of the circuits that control the cell cycle of the bacterium Caulobacter crescentus. Recent accomplishments include a systematic investigation of deletion phenotypes and substrate specificity of two-component signaling proteins in Caulobacter, leading to the discovery of a highly conserved, essential signaling pathway controlling cell envelope biogenesis and structure (Skerker et al. 2005), the delineation of a complete signaling pathway controlling stalk biogenesis during cell cycle progression (Biondi et al. 2006), and an integrated genetic circuit that ensures oscillations in activity of the master cell cycle regulator CtrA (Biondi et al., 2007).

Oliver Rando studied chromatin structure and function, using a novel microarray to determine nucleosome positions and modification states over 0.5 Mb of the yeast genome, at 20-base-pair resolution. This work has led to the following discoveries:

Most of the yeast genome is found in well positioned, rather than delocalized, nucleosomes; and most promoters have a long nucleosome-free region where transcription starts, which is enriched for poly-dA-dT stretches and conserved transcription factor binding sites (Yuan et al. 2005).
Putting conserved A-rich sequences at a heterologous location can make a nucleosome-free region (Raisner et al. 2005).
The pattern of histone modification in yeast nucleosomes is much less complex, and much less instructive, than advocates of a “histone code” have proposed (Liu et al. 2005).
nucleosomes exchange rapidly at promoters and at known chromatin boundary elements; histone replacement rates over coding regions correlate with polymerase density (Dion et al. 2007)

Christine Queitsch asked how genetic variation, maternal effects and chaperone levels determine developmental trajectories and fitness of Arabidopsis thaliana (Sangster et al. 2007). Her lab also developed an Arabidopsis genotyping array using insertions and deletion polymorphisms.

Hans Hofmann explored the molecular basis of neural and behavioral plasticity in the African cichlid fish Astatotilapia burtoni. His group used a combination of behavioral observation and microarray analysis (Renn et al. 2004) to map the genetic modules responsible for the dramatic behavioral and physiological transitions between territorial (“macho”) and non-territorial (“wimp”) males.