Mitochondrial dysfunction is a hallmark of several human diseases and somatic cell decline during aging. Germ cells must provide pristine, functional mitochondria to the embryo, although human oocytes can be stored decades before they are fertilized. C. elegans adults can survive for extended periods of time, during which they consume the gonad except for a few germ cells. Remarkably, these aged germ cells can regenerate the entire gonad. Here, I will use C. elegans to investigate the hypothesis that compositional differences between germline and somatic mitochondria underlie differences between these cell types. I will expand upon preliminary work showing that adult germ cell mitochondria possess a dramatically altered ratio of respiratory chain components compared to several somatic cell types. I propose three aims to test my hypothesis. I will first determine the point in development when germline and somatic mitochondrial become distinct, which is integral to understanding how such differences could be regulated. I will further examine the molecular basis of this difference, its relationship to mitochondrial activity and the consequences of altering these differences in both germline and somatic cells. Finally, I will test if engineering germ cell-specific mitochondrial characteristics in somatic cells can alter somatic function, potentially impacting the fate of cells subject to mitochondrial dysfunction during aging. These aims will uncover fundamental aspects of mitochondrial function during development and may reveal avenues through which the rejuvenating properties of germ cells could be applied to the soma.

Proteins can self-assemble to create hollow structures that encapsulate cargo. These appear in nature with various roles, such as reaction vessels for catalysis or viral capsids. Recently, the first examples of artificially engineered protein containers have appeared, presenting an exciting opportunity to develop synthetic assemblies to mimic the structure and function of their biological counterparts. The proposed research will develop a protein cage which can load nucleic acids into its core. This presents a solution to the delivery of therapeutic nucleic acids. A protein-based carrier has distinct advantages over current approaches, due to the high level of structural control and potential for site-specific modification. The encapsulation will be achieved through electrostatic interactions, exploiting the anionic nature of nucleic acids and directing them to the positively charged cavity of the protein cage. The first part of the project will focus on the design and expression of a hollow protein structure with a positively charged lumen. Secondly, the capacity for loading will be investigated, with a focus on the stability and capacity of the capsule to protect its cargo from degradation. Finally, the ability of the device to enter cells and regulate protein expression will be thoroughly investigated, including modification of the capsid surface to control cellular uptake and trafficking. This research will provide new insights into the potential of protein cages as active devices to regulate cellular function. Additionally, the resulting protein-DNA complexes provide an ideal model system for investigating the properties of naturally occurring protein compartments.

Understanding the mechanisms of longevity can provide ways to achieve healthy ageing. Naked mole rats (NMRs) are the longest-lived rodents that remain healthy until the end of their lives and are resistant to age-related diseases including cancer. One causative agent of this remarkable trait is the acquisition of high molecular mass hyaluronic acid (HMM-HA). The catalytic domain of the HA synthase gene HAS2 is highly conserved across several phyla, exhibiting two amino acid changes unique to NMRs that result in the synthesis of HMM-HA. I propose to create transgenic mice using CRISPR/Cas targeted genome editing that will contain the unique NMR AA changes engineered into the murine HAS2 gene. An orthogenic mouse model of the NMR HAS2 gene will tell whether HMM-HA can delay aging and prevent cancer in a species other than NMR. To identify genes involved in HMM-HA pathway and novel tumor suppressor mechanisms we will generate a whole-genome NMR CRISPR KO library. Primary NMR skin fibroblasts expressing SV40-LT antigen and Ras-V12 will be transduced with the library and malignantly transformed colonies will be selected. Unpublished data from the Gorbunova lab show that NMR hematopoietic stem cells (HSCs) are biased towards erythroid lineage. I will apply the NMR KO library to NMR bone marrow (BM) cells revealing genes responsible for the prevalence of erythroid development inherent to NMR HSCs. The genome scale NMR KO library will prove an invaluable tool for researchers utilizing NMR, while the discovery of new tumor suppressors and the molecular understanding of the longevity and cancer resistance of this remarkable rodent have great potential for biomedical applications.

Our goal is to unveil evolutionary and ecological roles of horizontal gene transfer (HGT) in natural ecosystems. HGT is believed to be promiscuous in nature and one of the major driving forces for microbial evolution. Yet, little is known about the frequency of HGT and its contribution to structures, interactions, and functions of natural microbial communities. HGT is often studied under artificial laboratory conditions using simple donor-recipient pairs of microbes. Furthermore, HGT events in natural communities are typically inferred from comparative genomics without experiments. Here, we propose to exploit the characteristic gut microbiota of honey bees as a versatile model to study the impact of HGT in a natural microbial community. Honey bees harbor remarkably stable and simple bacterial gut communities consisting of only eight bacterial species. Each species is represented by a number of divergent strains that can coexist. Preliminary genomic data suggests that HGT occurs within and across species in the community. In this proposal, we hypothesize that HGT confers robustness on the community level to hedge against environmental perturbations by two mechanisms: (I) HGT allows the community to maintain beneficial functions by increasing their redundancy, (II) HGT contributes to maintaining species diversity by spreading conditionally critical functions. To test these hypotheses, we will combine genomics, experimental, and theoretical approaches and investigate dynamics, functions, and consequences of HGT in the bee gut microbiota. Using in vivo screening and population genomics, we will first establish a comprehensive gene catalogue of horizontally transferred elements. We will develop an ex vivo system that will enable us to track HGT and community structure changes at the single-cell resolution in real-time in the actively maintained bee gut. Genomics and ex vivo data will be used to develop theoretical and computational models of eco-evolutionary dynamics that account for how HGT alter quantitative ecological interactions. Ultimately, we will test HGT-deficient communities of bee gut symbionts in bee colonization experiments to show the influence of identified HGT on community structure and symbiotic functions in vivo. The project will lead to novel insights into eco-evolutionary dynamics by incorporating HGT into community dynamics.

Cell fate is determined by epigenetic controls that determine patterns of gene expression. Differentiated cells are typically locked into stable steady states of restricted expression, while in cancer cells and stem cells restrictions are diminished or lost. Gene expression is determined biochemically, by transcription and chromatin remodeling factors, and physically, by access restriction or enhancement. Here, I focus on the largely unexplored physical control of gene expression. I test the hypothesis that non- equilibrium stress fluctuations in the nucleus serve to locally enhance transport and DNA accessibility. I will probe in how far enhanced fluctuations and decreased packing density in cancer cells and stem cells are directly correlated with enhanced and possibly pathologically uncontrolled transcription activity. I will develop and use new methods to track fluctuations in the nucleus of control cells and cancer cells with near infrared fluorescent single-walled carbon nanotubes (SWNTs). These extremely photostable and precisely targetable nano-probes allow us to track dynamics with millisecond resolution for up to hours. I will apply and further develop a new super-resolution microscopy technique known as Double-Helix Point Spread Function Microscopy (DH-PSF), which uses an axially extended helical point spread function to achieve simultaneous extended z-resolution and orientation tracking of multiple SWNT probes in 3D. I will use “active matter” concepts from statistical physics to analyze non-equilibrium and equilibrium fluctuations and measure local chromatin mechanical response, as well as transport through the chromatin network.

Cell reprogramming provides a potentially unlimited cell source for in vitro tissue engineering for biomedical research and regenerative medicine. Despite significant advancements in the field, it remains elusive whether the reprogrammed cells fully resemble their native counterparts. This is mainly because we lack understanding of biological pathways in single cells. The recently developed Fluorescent In Situ Sequencing (FISSEQ) allows for in situ sequencing of RNAs of single cells in tissues without pre-selection of the target RNA population. By performing in situ sequencing of DNA barcode-linked antibodies that specifically bind to target proteins, the FISSEQ method can be adapted to perform in situ proteomic study. This project aims to expend and modify the RNA FISSEQ technique to characterize chromatin-associated proteins that are important for transcriptional regulation at single cell level. We will characterize these chromatin-bound proteins in single cells of both native and lab-generated human cerebral tissues. By identifying the key regulatory factors for cell-fate determination, we will be able improve our ability to guide engineering of physiologically-relevant tissues.

Cells in developing organisms can keep track of time. They use this to make important decisions - for example on when to turn or to stop in the case of migrating axons or cells - without being instructed to do so by signals from other cells in their surroundings. But how such internal clocks work, and importantly, how they are made to be so precise, is still largely unknown. Previous work by one of the participating teams has shown that during the development of the nematode worm C. elegans, the migration of a neuroblast is regulated through the timed expression of a signaling receptor. This system provides a powerful assay to study at single cell level how an internal clock controls gene expression. The three teams will use a unique combination of genetics, evolutionary biology and mathematical modeling to gain detailed insight into the workings of this timing mechanism. They will investigate how timing is mediated at the transcriptional level, how robust this is to environmental variations and how this mechanism has evolved in other nematode species. Importantly, these results will be used in mathematical modeling to gain insight into the underlying regulatory architecture and to make predictions that will be tested in further experiments. Such interplay between experimental and theoretical analysis is a powerful and innovative approach that will enable the three teams to gain deep understanding of how cells measure time.

Vocal learning is a fundamental trait of spoken language and yet the neuro-molecular mechanisms underpinning this trait are poorly understood. The capacity for vocal learning has only been identified in a handful of non-human mammals; cetaceans, elephants and bats. The first evidence for some vocal adaptations in non-human primates is only now emerging. Despite intense interest, these limited experimental options mean that the neural underpinning of mammalian vocal learning has been massively understudied. Some birds, such as parrots, hummingbirds and songbirds, are also vocal learners. Work in songbirds has provided valuable insights, but the evolutionary divergence of avian brain structure is a barrier for direct extrapolation to mammals. Furthermore the lack of a mammalian animal model means that it is unclear whether or not mammals and birds use similar mechanisms during vocal learning. As a result, fundamental questions in the field remain unanswered, such as: how is vocal learning encoded in the mammalian brain, how do neurological and genetic substrates contribute to this trait, and are these mechanisms similar across divergent species?
We will establish bats as an ideal model system because they are mammals that have robustly shown vocal learning and have remarkable capacity for combining vocal phonemes to create new types of vocalisations. Bats can be maintained in laboratory colonies, and there is already a wealth of information regarding the neuroethological mechanisms by which they produce and perceive their vocalizations. We aim to establish bats as the first mammalian model of vocal learning by designing innovative communication paradigms coupled to comprehensive neurological and genetic interrogations. This proposal integrates psychophysical, anatomical, electrophysiological & genetic research in an innovative way to begin to address the encoding of vocal learning in the mammalian brain.
This complementary and coordinated research effort, focuses on a tractable and very promising animal model, addresses a fundamental gap in the field and is likely to significantly advance our knowledge about the origins of vocal learning and ultimately human speech. We believe that this work will have important repercussions across sensory ecology, neuroscience, genetics, evolutionary biology and linguistics.

The phenotype of an animal cannot be explained entirely by its genes. It is now clear that factors other than the genome contribute to the development and dynamic homeostasis of animals. Two fundamentally important factors are epigenetic regulations and the microbial communities associated with the animal. Unlike the genes and regulatory regions of the genome, epigenetics and microbial composition can be rapidly modified by environmental cues, and may thus represent mechanisms for rapid acclimation to a changing environment. At present, the individual functions of epigenetics, microbiomes, and genomic mutations are largely studied in isolation, particularly for species in marine ecosystems. This leaves significant questions open for how these mechanisms intersect in the acclimation and adaptation of organisms.
The aim of this research is to determine how epigenetic regulations and microbial communities participate in thermal acclimation of a coastal marine species residing in a dynamic temperature environment, and how these non-genetic factors interact with each other. The model species used for this study, the sea anemone Nematostella vectensis, enables us to carry out unprecedented functional experiments to dissect the interactions between microbes and epigenetic mechanisms in the acclimation of the holobiont.
We will first monitor the physiological, transcriptomic, epigenetic and microbial changes associated with thermal acclimation. We will then separate the effects of microbial and epigenetic effects in a series of bacterial transplantation experiments. Finally, we will carry out gene knockdown and over-expression experiments to elucidate the function of host genes in epigenetic regulations and the plasticity of the microbiota. We hypothesize that changes in the microbial community improve the thermal tolerance of the host, and the epigenetic landscape is responding both to shifts in temperature and altered microbial composition.
Studying these interactions will require a concerted effort drawing on the diversity of expertise across our three research groups. Our results will have important consequences for our understanding of the response of marine species to climate change, and more broadly they will shed light on many unanswered questions regarding the role of epigenetic regulations and microbes in animal ecology and evolution.

Mouse and human mutations in the RNA-binding protein Bicaudal-C (Bicc1) are associated with cystic kidneys that are reminiscent of hereditary polycystic kidney diseases (PKD) (Cogswell et al., 2003; Kraus et al., 2012; Maisonneuve et al., 2009). Direct mRNA targets of Bicc1 include adenylate cyclase 6 (AC6) and PKA inhibitor (PKIα), members of the cAMP/PKA signaling pathway (Piazzon et al., 2012), and Ddx5, a gene encoding a RNA helicase of unknown renal function ((Zhang et al., 2013) and unpublished results). In addition, Bicc1 stimulates the transcription and translation of polycystin-2, although these effects are likely indirect ((Tran et al., 2010), and unpublished results). More recently, the Constam lab found that endogenous Bicc1 in kidneys also binds the protein Inversin (Invs), which is mutated in nephronophthisis type 2 patients (Florian Bernet, PhD thesis, unpublished). How this complex is de-regulated and its role in cystic diseases remains obscure. The objective of this project is to develop molecular biology tools to characterize the regulation of Bicc1 activity by Invs and calcium signaling. Novel fluorescent and affinity-tagged versions of Bicc1 and their targeted insertion at the endogenous locus will be used for live imaging experiments, co-localization studies and co-purification of interacting factors to investigate how the de-regulation Bicc1 silencing activity triggers cystic disease. Outcomes of the proposed experiments will help to improve our knowledge of cyst development and to explore new therapeutic avenues.

The acquisition of novel traits is a central pillar in evolutionary biology. Adaptive radiation is recognized as a major process that leads to a rapid diversification of the progeny of a single ancestor into a number of functionally different, genetically isolated populations. Adaptive radiation is proposed to be driven by rapid adaptation to new or underutilized ecological niches that leads to phenotypic diversification and eventually to speciation. Experimental evolution of unicellular prokaryotes can produce adaptive radiation in a controlled environment, but this process and its population dynamics and genetic basis has not been studied in eukaryotes. Here, I propose a novel approach for experimentally inducing adaptive radiation in a eukaryotic model organism (S. cerevisiae). By evolving parallel yeast clones in the simultaneous presence of several, different nutrient sources that cells cannot exploit, I aim to evolve the ancestral form into separate populations, each occupying a different, novel metabolic niche. By barcoding lineages and exploiting next generation sequencing techniques, I plan to reveal the population dynamics and adaptive mutations that allow cells to evolve new metabolic traits. I will evolve both haploid and diploid cells to test the influence of genomic architecture on the radiation process; for example do loss of function mutations drive radiation of haploids, whereas gain of function mutations predominate in diploids, and does niche specialization alter mating preference or reduce the fitness of the products of inter-population matings?

Humans and monkeys are very good at interpreting intent from other’s body movements and facial expressions. This visual recognition is achieved seemingly effortlessly. This visual recognition is also highly accurate, even though the possible number of body movements and facial expressions is extremely large. While much is now know about the different brain mechanisms involved in the recognition of objects, scenes and faces, little is known on how the visual system recognizes other agent’s intent. We will use functional Magnetic Resonance Imaging (fMRI) in humans and fMRI and single cell recordings in monkeys to identify the neural mechanisms responsible for this recognition. The neuroimaging data will be analyzed using sophisticated machine learning algorithms. These methods of analysis will be used to identify the regions of interest (ROIs) involved in this visual recognition of intent, determine their interaction, and study the image or higher-level features used in each ROI. We hypothesize that, as in object and face recognition, there is a hierarchy of areas with increasing level of abstraction. Finally, we will use the results of our experimental data to derive a neurophysiologically plausible model and optogenetics to test for the causal role of these computations.

Synchronization in connected populations is routinely encountered in a variety of situations ranging from epidemic outbreaks to species extinction. This research proposal aims to explore the effect of nonlocality and heterogeneity, which are hallmarks of real world ecological and epidemiological networks, on the conditions for synchronization. The proposed research aims to employ experiments as well as computer simulations to better understand the emergence of synchronization in networks of connected populations. The chief goal of the proposed computational model is to construct a ‘phase diagram’ which demarcates regimes corresponding to the presence and absence of synchronization in a three-dimensional parameter space comprising of the migration rate, the degree of nonlocality and the degree of heterogeneity. The second aim is to elucidate the nature of the transition from synchronous to asynchronous behaviour as one of the three parameters is changed. The proposed experiments aim to employ a model bacterial system recently developed in the Gore lab to directly test the model’s predictions. The experiments involve simultaneously growing bacterial cultures and periodically measuring population densities using spectrophotometry and flow cytometry. Migration is facilitated by mixing a fixed fraction of a given culture with its neighboring cultures. Nonlocality and heterogeneity can be tuned by choosing appropriate neighbors for mixing. The extent of synchronization can be quantified directly from the population time series data. Results from these experiments and simulations will be of direct relevance in organizing conservation as well as vaccination programmes.

Curiosity, defined as the intrinsic desire to know, is among the last unexplored frontiers of higher cognition, and we know very little about its neural mechanisms. We propose to address this question by developing a program for studying curiosity using an integrated empirical/computational approach in humans and non-human primates. We examine the hypothesis that curiosity is a family of mechanisms that evolved to allow animals to maximize their knowledge of the useful properties of the world – i.e., the regularities that exist in the world - using active, targeted investigations. In two experiments, we probe two processes that contribute to curiosity-based exploration. In experiment 1 we examine how agents ascribe value (“interest”) to surprising events, by (1) developing a new behavioral task where children and monkeys make tradeoffs between exploring for sources of reward versus exploring a surprising item, and (2) investigating single-neuron responses related to curiosity-based exploration in cortical areas implicated in the control of attention. In experiment 2 we ask whether subjects show a more sophisticated form of curiosity guided by learning progress – a meta-cognitive tracking of the amount of learning that the individual can make in a task - using new behavioral paradigms where children or non-human primates see a set of symbolic learning problems and freely choose which ones to explore. To quantitatively model curiosity we use the framework of Bayesian Reinforcement Learning, which allows us to infer the agent’s beliefs about the probabilities of various events (rewards, surprises, or learning), the value they place on sampling different events, and how this value depends on the task context. The studies closely integrate the expertise of the 3 member teams in developmental psychology (Kidd), neurophysiology of non-human primates (Gottlieb) and computational modeling of active learning in robotic systems (Oudeyer). Our goal is to develop an integrated theory that (1) incorporates curiosity in established quantitative frameworks of learning and decision making, (2) links it with core cognitive functions such as selective attention, (3) compares its expression in humans and non-human primates, and (4) begins to elucidate its neural mechanisms.

In the brain, the balance between excitation and modulation plays a crucial role in numerous cognitive processes and pathological impairments of this balance may underlie several neurological disorders. Indeed, dysregulations of two neurotransmitter receptor families, glutamate NMDA (NMDAR) and dopamine (DAR) receptors, are believed to play a direct role in the emergence of psychotic disorders such as schizophrenia. Thus, understanding how these two neurotransmission systems communicate is of growing interest. Intensive investigations have identified intracellular and genomic cascades involved in this dialogue. However, the recent discovery that NMDAR and DAR diffuse and physically interact at the surface of neurons profoundly changed our view of neurotransmitter signaling. A major challenge is now to unveil the contribution of DAR/NMDAR complexes in the regulation of living brain circuits. To this aim, we will combine complementary expertise to investigate the surface crosstalk between NMDAR and DAR both in physiological and psychotic disorder contexts, focusing our attention on D1R, D2R, and NMDAR in hippocampal, prefrontal and striatal circuits. We will assess the roles of this dynamic receptor interplay from a multi-scale perspective. We will first investigate the rules of the DAR/NMDAR surface dialogue at the single molecule level by mapping all interaction sites and monitoring receptor distribution and diffusion using multi-color single molecule imaging in vitro and in brain slices at several developmental stages, both in control and in models of schizophrenia. We will then investigate whether the DAR/NMDAR surface crosstalk regulates glutamatergic transmission along the dendritic tree, its impact on NMDAR-dependent plasticity, and its role in dendritic outgrowth and spine maturation during development using two-photon calcium imaging, electrophysiological recording and competing peptides. Finally, we will investigate the influence of the DAR/NMDAR interplay on physiological and pathological network dynamics and behaviors using multi-electrode array recordings, optogenetic release of dopamine, competing peptides and “data-mining” statistics. In conclusion, we will join international forces to shed the first lights on this unexpected dynamic integration level using nanoscopic, macroscopic and mesoscopic approaches.

The oceans are teeming with microbial life, which impacts global biogeochemistry, climate and human health. Omics methods enable us to identify "who is there" and their genetic potential, but understanding how these organisms interact in nature and how they affect biogeochemical processes remains a major open challenge. A crucial component of this puzzle, at the heart of the cycling of nutrients in the biosphere, is the coupling between organisms that fix carbon using solar energy (phytoplankton) and microbes that rely on organic carbon for energy and growth (heterotrophs). The diversity of microbial species and the richness of their metabolism make the problem of predicting how each pair of species will interact impossible to address using traditional approaches. We propose to tackle this challenge through a tightly integrated combination of genome scale modeling and laboratory experiments, to identify genomic traits dictating how environmentally-relevant microbes interact. We will generate a library of ~100 heterotrophic bacteria representing major marine lineages and fully sequence and annotate their genomes. We will implement genome-scale (dynamic flux balance) models of each of these organisms, as well as of four key phytoplankton species, calibrating these models using high-throughput measurements of growth under different conditions. We then grow binary phytoplankton-heterotroph co-cultures in the lab and compare the results with in-silico models of the interacting genomes, studying key interactions in more detail. Recognizing that microbes in the ocean are seldom in steady-state, our work will encompass exponential growth, stationary stage and culture death. Most importantly, missing or incorrectly predicted interactions will give us an opportunity to revisit our models, and suggest the mediation of alternative processes such as allelopathy or other types of chemical signaling. This study will provide the first detailed "roadmap" linking genomic traits (genes and metabolic pathways) and rate measurements with species interactions in environmentally-relevant microbes. Our approach, which - we anticipate - will be embedded in a not-so-distant future in global-scale models of the Earth System, will provide a critical stepping-stone towards predicting how marine microbial systems will evolve in a changing world.

Monomeric actin is readily detectable in the nucleus of mammalian somatic cells. Whether this pool of actin forms filaments has been debated for decades. Recent advancements have established that, indeed, nuclear actin can assemble dynamic filamentous structures in response to external stimuli. However, how nuclear actin filaments exert their biological function remains an unanswered question in cell biology. This is particularly relevant given the highly dynamic nature of nuclear actin filaments, which reside within the intricately organised nucleus that harbours the genome in the form of chromatin. Thus, through this collaboration, we aim to address this by developing and applying novel approaches for optogenetics, advanced cell-imaging and next generation sequencing. Accordingly, we will employ tools to spatiotemporally control nuclear actin filaments with light, and determine their role in regulating chromatin organisation and nuclear structure. Further, we will determine the cell and biological consequences of actin filament-driven chromatin organisation/re-organisation in relation to cell division and cellular reprogramming. By completing these studies, we envisage to advance our understanding of the biological functions of nuclear actin filaments in a manner that will encompass nuclear actin polymerisation as a key mediator of chromatin organisation and regulation. Beyond this, findings from our studies may have translational potential in areas of regenerative medicine, and may delineate new mechanisms that underlie ageing and degeneration.

Retinotopy, the pattern of axonal projections in visual processing centers that preserves the spatial arrangement of photoreceptor (PR) in the retina is a key feature of any complex visual system. I offer to investigate how retinotopy is built in the Drosophila visual system focusing on the medulla neuropile. Medulla neurons are generated sequentially by the Outer Proliferation Center and build 800 columns onto which inner PRs (R7 and R8), also sequentially generated along a posterior to anterior axis in the eye disc by the progression of the morphogenetic furrow, project retinotopically.

To find what are the medulla neurons that support PR retinotopic projections, I will first generate tools to visualize medulla column generation and test by a genetic approach several neuronal populations. I will then determine whether developmental synchrony between the waves of differentiation in the eye disc and medulla is required along the antero-posterior axis for PR retinotopic projection onto medulla columns. I will genetically modulate the speed of both waves in order to desynchronize them and look for retinotopic defects. I will then investigate how those two waves are coordinated. In parallel, I will find genes responsible for dorso-ventral patterning of the medulla by a differential transcriptomic strategy. I will purify ventral medulla neurons by intersectional approaches and compare their expression profile to the entire medulla neuron population.

Completing those three aims will allow me to uncover the developmental rules and signaling pathways responsible for building a complex and immensely precise architecture such as neuronal retinotopic maps.

Enhancers and core promoters are key regulatory elements underlying the precise spatio-temporal regulation of gene expression. Despite recent efforts to map and characterize these elements genome-wide, it is still unclear how their activities are encoded in the genome. Moreover, we are just beginning to unravel the role of the specificity between enhancers and core promoters in transcriptional regulation.
Novel genome-wide activity-based assays for identification of enhancers and core promoters (STARR-seq and its variant) developed in the host lab provide a unique opportunity to quantify sequence-encoded activities in a highly controlled manner. In collaboration with the experimentalists I propose to apply STARR-seq in human cells to measure both enhancer and core promoter activities and assess their specificity. Using these data I will apply sequence analysis and modeling approaches with an aim to establish a quantitative relationship between sequence and activity for both types of elements. I am aiming to reveal differences and commonalities in sequence requirements for a genomic region to act as an enhancer or as a core promoter, allowing me to build a comprehensive model of transcriptional regulation in human. Given the diversity of mammalian core promoters, I will also test whether enhancers and core promoters exhibit mutual preferences and will attempt to uncover sequence determinants underlying such specificity. Finally, using published datasets I will assess how these intrinsic activities are modulated by the endogenous chromatin and sequence context. My results will further our understanding of gene regulatory elements and their interpretation by human cells.

Like human babies learning speech, young songbirds learn to produce song by imitating adults. Both babies and young birds need to listen to adult vocalizations, extract the relevant information and store it in the brain and then improve their own ‘babbling’ sounds step by step towards the adult version. Experiments on vocal learning have focused mainly on sounds, although sight also seems to play an important role. Babies of only 8 weeks of age already associate speech sounds with the correct mouth position required to make that sound. Moreover, people report that they hear the syllable ‘ga’ when in fact a loudspeaker plays ‘pa’ while a video screen shows the face of a speaker mouthing ‘ka’. This perceptual phenomenon is known as the ‘McGurk' effect. It reveals that sensory information from the ear and eye are integrated in the brain to form a combined, multimodal percept.
Like human speech, birdsong can be heard and seen. Young birds see their fathers moving their beaks while hearing their song. There is substantial evidence to assume that the combined exposure to acoustic and visual cues enhances song learning not only in human but also in bird babies. We aim to uncover the role of this multimodal perception and its underlying neural architecture in vocal learning. Focusing on the zebra finch, the prevalent animal model for molecular, neural and developmental aspects of human speech we will adopt a robotic approach that allows full experimental control over auditory and visual cues.
Simultaneously recording acoustic and visual cues of singing adult birds, we will analyze synchrony between different song elements and the opening and closing of a bird's beak. We will use this to develop a robotic zebra finch that allows us to create matched and mismatched multimodal stimuli. Juvenile birds will be raised with various tutoring regimes to assess the relative roles of unimodal versus multimodal cues, life versus robot cues, and matched versus mismatched cues. Upon reaching adulthood, tutored birds will be tested on their song copy performance (males only), song preference (females only), or discrimination ability (all birds). Finally, we will use a subset of tutored birds to study underlying neural processes by using molecular tools that allow us to pinpoint where in the brain multimodal information is processed and how this is linked to vocal learning.