Perceiving and interpreting our sensory environment feels subjectively effortless, yet, it is one of the computationally most challenging tasks the brain needs to solve. Critically, our brain constantly fuses current sensory input with expectations derived from prior experience in order to be able to reliably interpret our environment in spite of ever-changing conditions, such as lighting, pose, or configuration, making every single experience unique. That prior expectations are constantly shaping our percepts has been well known for over a century from behavioral studies, but how neural circuits in sensory areas of the brain achieve this sensory-prior fusion has remained largely unknown. Progress in several research fields make it possible now to address the neural bases of sensory-prior fusion: novel recording techniques provide an unprecedented opportunity to track the activity of large populations of neurons over extended periods of time; advances in artificial intelligence and machine learning provide powerful new tools to analyse complex data; and advances in computational theories allow us to formulate experimentally testable hypotheses and compare their predictions to data. These techniques are now being used to great effect in sensory neuroscience, but almost exclusively to characterise how sensory input alone impacts neural responses along different stages of the visual processing system of the brain. Our project aims to bring these advances to bear on understanding the effects of prior expectations on neural responses, and their fusion with sensory information. We will rely on complementary approaches that provide insights into different aspects of sensory-prior fusion: experts in mouse experiments and monkey experiments will train complex tasks and record from brain areas that contribute to sensory-prior fusion, while experts in machine learning-based data analysis and computational neuroscience will develop cutting-edge analysis techniques and models to guide experiments and interpret data in a formal theoretical framework. Our project will thus help us understand how a brain that is composed of unreliable components and prone to a variety of perceptual distortions and illusions can efficiently process rapidly changing, cluttered, and noisy scenes to achieve feats such as driving us home safely in busy traffic.

Understanding how out of single cells functional tissues and organs develop is a major challenge of biology. Recent progress allows us to grow organ-like cell assemblies (organoids) from stem cells in vitro. Organoids offer great potential for studying diseases and development. However, in many cases we do not yet understand how these complex tissues emerge out of progenitor stem cells. A common feature in the initial growth phase of many organoid systems is the formation of a polarized epithelial cyst with a single or multiple internal apical lumen. This initial transition into an epithelial cyst establishes a tissue template that on the one hand enables maintenance of progenitor/stem cells (niche) and on the other hand guides the patterning of differentiated cells into a functional tissue. Our aim is to understand how the interplay between proliferation (cell divisions), polarization (epithelial transition) and differentiation (patterning) leads to self-organization of this epithelial progenitor template and how this structure facilitates correct patterning into functional organoids. To this end, we will systematically control and characterize the early growth phase of two organoid systems (pancreatic and neural tube) using microfabrication and micro-patterning approaches. We will quantify evolution of cell shapes, adhesion and cortical forces, apical-basal polarization and differentiation as a function of initial cell contact patterns. This approach will provide the means to find rules how local cell interactions (cell-cell, cell-matrix, cell-lumen) are connected to tissue growth and differentiation. We will then test sufficiency of the hypothetical rules to generate the observed organoid structures using an in silico mechano-chemical model. Taken together, by dissecting the early growth phase of two organoid systems, we aim to uncover the common rules on how progenitors establish a polarized epithelial template, and how this template is then differentially used to generate organ specific differentiation patterns.

Unlike humans, some multicellular organisms are able to regenerate their organs after severe injuries. Understanding how regeneration properties emerge from fundamental biological processes such as gene expression and stem cell differentiation is of utmost importance for basic science and regenerative medicine. Nevertheless, little is known on the mechanisms underlying the initiation, maintenance and control of regeneration. Therefore, many fundamental biological questions remain unanswered, i.e.: (i) how do stem cells migrate, divide and differentiate? and (ii) how do regenerating organs control their size and orientation? Thanks to their astonishing regenerative properties and simple body plan, planarians form an ideal model system for the study of regeneration. However, due to their opaque tissues, time-lapse microscopy of living planarians has so far been impossible and a comprehensive understanding of the mechanisms underlying their regeneration is still lacking. In the proposed research, I aim to use an interdisciplinary approach to characterize, for the first time, the single-cell dynamics of planarian regeneration. I will use light sheet microscopy, an ideal technique for live imaging of large specimens, and design microfluidic devices to keep the organism at physiological conditions. Fluorescently labelled animals will be used to follow the regeneration dynamics at the single-cell level. Together with complementary genetic and molecular biology techniques, this approach will give us new insights on single stem cells and gene expression dynamics during regeneration.

Life on earth depends upon the photosynthetic conversion of H2O, CO2 and light into fixed carbon (C) compounds including polysaccharides and lipids, and yet the biosynthesis, regulation and many aspects of the function of the photosynthetic apparatus are still not well understood. A major challenge for photosynthetic organisms is to capture light energy for growth while at the same time eliminating excess absorbed excitation to prevent photodamage, which can be energetically costly and potentially lethal. A major photoprotective mechanism in plants and most algae is qE (quenching energy), a quenching mechanism that dissipates absorbed excitation energy as heat, ensuring survival even under adverse conditions. In the green alga Chlamydomonas reinhardtii (Cr), two LHCSR proteins (Light Harvesting Complex Stress Related; LHCSR1 and LHCSR3) are required for qE function.
The LHCSRs are nucleus encoded, and their expression is governed by transcriptional and post-transcriptional processes that are impacted by CO2, nutrient availability, photosynthetic electron transport (PET), blue light perception and calcium(Ca2+)-elicited signalling. Despite their fundamental importance, mechanisms that regulate LHCSR accumulation are still not well understood and complicated by multiple inputs. For example, our findings demonstrate that LHCSR control is impacted by the metabolism of fixed C compounds. The combined use of physiology, biochemistry, genomics and mathematical modeling will allow us to elucidate transcriptional, post-transcriptional and epigenetic processes that link photosynthesis, C metabolism and photoperception to qE activity, and help establish a systems level view of photosynthetic dynamics.

In nature, cellular tissues efficiently remodel to attain crucial biological processes, both in health (organogenesis and development) and disease (cancer invasion and metastasis). Such reshaping events require collective cell migration, by means of which tissues can generate forces at large length scales. Although studies in vivo have evidenced the essential role of collective migration modes and flows within reshaping tissues, in vitro studies are necessary to explore these processes in simpler and better controlled microenvironments. In particular, this project aims at inducing reshaping events within confluent monolayers of cells by controlling cell flows through patterns of differential adhesiveness. First, we will explore the effects of geometrical confinement in the steady state positioning of topological defects in cellular nematics as a tool for localizing stress singularities of different strengths within cell tissues. Second, we will gain control over cellular active flows in order to induce unidirectional collective migration patterns. To this purpose, we will use gradients of surface-bound cues to control cell migration (haptotaxis) that, together with geometrical constraints, will provide with the tools for controlling flow and force fields within tissues. Finally, we will design patterns combining geometrical confinement and adhesion gradients to locally frustrate active flows and induce the creation of folds in preassigned regions within proliferative cell monolayers. Our work not only will provide experimental verification of the link between tissue mechanodynamics and reshaping, but also envisions the development of living 3D tissue structures in vitro.

The indeterminate growth of plant meristems is mediated by the long-term maintenance of stem cells. While patterns like the placement of stem cells are highly dependent on cell-cell communication in plants, there is incomplete understanding of the sources of signals that pattern the meristem. In addition, there is even less knowledge about how stem cells form during organogenesis and regeneration. In plant roots and other organs, patterning processes have been shown to be highly dependent on symplastic signaling through plasmodesmata -- membrane lined channels that connect every plant cells and allow trans-cellular movements of proteins (including transcription factors), miRNAs, RNAs, sugars and ions. To study cell-cell signaling on a broad scale, the host institute has developed a novel system to monitor complex communication from one cell to another during development. The system relies on recently developed tools to inducibly block plasmodesmata combined with single-cell RNA-seq profiling in plants. Here I propose to use this new set of techniques to sequentially block different cell identities and assess the consequent transcriptional changes using single-cell RNA-seq – a strategy we call “Block-Seq.” In particular, single-cell resolution offers the advantage of detecting responses in a subset of cells rather than expecting a tissue to respond uniformly. For the first time, these approaches will allow me to map communication from one cell type to another in the root meristem.

Although cell division is fundamental to malaria parasite proliferation its mechanisms are completely understudied. Plasmodium falciparum division, called schizogony, displays striking differences when compared to any model organism suggesting the presence of non-canonical division pathways. Key specificities of schizogony are the absence of classical mitotic checkpoints and asynchronous nuclear division. The number of emerging daughter cells is highly variable and plasmodial centrosomes have atypical structure and dynamics. Further, the parasite undergoes closed mitosis, which requires the nuclear membrane to be split between dividing chromosomes by a mechanism that is entirely elusive. Studying the dynamics of schizogony requires an exquisite temporal and spatial resolution, which could only be achieved in recent years. I propose a combination of cutting-edge microscopy and genome-editing to uncover molecular events underlying atypical parasite division mechanisms. Using live cell, super-resolution, and correlative imaging I want to generate a robust cell biological framework describing key division events. Proteomics and genome editing will enable me to functionally characterize the plasmodium-specific centrosome. Nuclear membrane fission will be investigated by correlative light and cryo-electron microscopy. These ambitious and novel approaches will drastically improve our understanding of malaria parasite proliferation. They will provide insights into diversity of cell division mechanisms beyond what has been studied in model organisms. Investigating schizogony will open up new intervention strategies against malaria, which remains a major public health issue.

Structural colors using guanine crystals are widespread in nature and can be found in many organisms. While the physical basis for producing structural colors is well established, the cellular processes underlying the formation and regulation of these structures is poorly understood. We propose to study these intriguing, previously unexplored, cellular processes using the tunable structural colors of the zebrafish as a model. The structural colors of the zebrafish offer a unique opportunity due to the combination of tunable structural colors, vast information about the genome and access to developing larvae. To study these cellular processes I will join the Lab of Jennifer Lippincott-Schwartz, where we will use state-of-the-art super resolution imaging techniques and molecular cell biology tools, which are well established in the Lippincott-Schwartz lab together with cryo-EM, micro-spot X-ray diffraction and crystallography, in which I have gained expertise during my Ph.D. studies. This research will yield new insights on both the transportation and trafficking of nucleobases within the cell and will uncover new paradigms in biologically controlled organic crystal formation. Our findings will also provide new information on the interface between organic crystals and the cell cytoskeleton and may also inspire new therapeutic approaches for pathological conditions arising from uncontrolled organic crystallization such as gout and urate kidney stones. This research represents a radical change in direction for me. It will provide me with new skills in molecular-cell-biology, biochemistry, and experience in state- of-the-art super resolution fluorescence microscopy.

We know coffee by its taste, but also by its name and association with caffeine. Studies of brain activity indeed confirm that the brain simultaneously processes stimuli across distributed areas, each selective for a specific stimulus aspect. However, the study of stimulus-representations, which is crucial for understanding the computational principles underlying brain activity, is currently constrained to the detailed inspection of isolated brain areas. I propose a new formalism and method that may allow studying representations as a distributed neural process, by employing novel brain imaging, stimulation and computational research approaches. I first wish to understand how different stimulus-aspects are represented in distributed brain areas. In line with existing theoretical accounts, I aim to demonstrate that each brain area forms representations that reflect the natural world statistics of the information it processes, and this principle can be used for linking distributed representations of a single stimulus. Next, I will test whether parallel neural representations co-express to bind different aspects of a given stimulus. Yet, to enable adaptive interaction with the environment, behaviorally-irrelevant representations must be inhibited. I aspire to support this conjecture experimentally by showing that behaviorally-irrelevant representations emerge when inhibition is alleviated. This line of work promises insight into computations that underlie distributed cognitive processes, by suggesting a principle for understanding the relationships between distributed stimulus-representations, and a mechanism that regulates their expression to allow flexible behavior.

Biology has become a data science. The vastness of data acquired in life science require advanced analysis methods and the demand for computation and mathematical modeling is higher than ever before. In this circumstance, I am proposing a new framework in deciphering the information flow within molecular pathways utilizing the financial market analysis skills I acquired during my PhD. The current approach to studying complex molecular pathways in cell biology is to perturb pathway components and then classify the effect as a phenotype. This approach has enabled cell biologist to discover the molecular players of pathways and even to infer hierarchical interaction cascades in pathways. However, under conditions of nonlinearity and high redundancy, the interpretation of these phenotypes can become exceedingly complicated. The proposed research will focus on the following 2 aims. Establish the mechanisms of information flow within the GEF, GTPase network. Establish the mechanisms of detecting the transient topology with protrusion events. The Significance of the proposed research derives from; 1) the project challenges the experimental paradigm of molecular cell biology in that it seeks to circumvent the limitations of molecular perturbation in deciphering pathways with nonlinearity and redundancy; 2) the project introduce to the field of molecular cell biology a theoretical foundation for the definition of causality; and 3) the unique data sets available between the Danuser and Hahn labs will allow me to develop this framework on the example of GEF-GTPase interaction networks, in which the current tools fail to address the redundancy and non-linearity.

Is our perception of time and distance the same when we walk in the park, travel in a high speed train, or wait for the starting gun in a field race? Over a century ago, Einstein opened the way into considering space and time as relative variables. New evidence now suggests that areas of our brain might have been doing so all along. In this case, however, rather than resulting from the gravitational pull of a massive star, the effect would be the natural consequence of a simple principle of economy. Not all dimensions need to be represented all the time; one could possibly do better with simplified schemas adapted to every situation in an information compressed manner.
The entorhinal cortex is an area of the brain where a rodent or human represents its own position in space and time. Recent evidence suggests that spatio-temporal dimensions can actually be represented in a flexible way, depending on behavioral demands. We lack, however, all knowledge about the repertoire of possible schemas that it can choose from, on how these schemas are formed and stored, and on the neural mechanisms behind the selection process. We here propose to study the neural basis of this “brain relativity” in rodents by means of a variety of tools that include behavioral experiments, recordings of neural activity, manipulation of specific neural circuits and computational modeling. We will challenge animals by making them explore artificial environments with geometries that require a switch between representations, as would happen when touring through an Escher scene. A room will be slowly morphed into a corridor, or distance estimation into time. We hope that this will allow us to understand the shape that these representations have in the brain, their number, and the mechanism that controls which one is applied at any given time. We also plan to identify specific circuits of the rodent entorhinal cortex that encode information about space and time using state-of-the art genetic tools, which will allow us to gain on-line control over the selection of schemas and its behavioral consequences. Along the process, results will be compared with computational models to understand the logic underlying each observation. All this will help us to get a deeper insight into how we perceive and interact with space and time in our daily experience.

Directed cell migration is at the core of vital processes in immune response or development. We know key mechanisms of 2D migration, especially how remodeling the actin cytoskeleton forms leading/trailing edges. 3D migration deviates considerably from 2D behavior, is physiologically more relevant and barely understood, mainly since available imaging fails to provide sufficient spatial and temporal resolution but affects living specimen adversely. I propose a novel microscope to provide 3D subdiffraction resolution at several frames per second, while being “gentle” due to low intensities. I strive to combine benefits of light sheet microscopy and coordinate-targeted nanoscopy with protected STED: A secondary off-state protects emitters from state-cycling and photobleaching during depletion. Selective illumination of the focus-plane and widefield detection limit phototoxicity at high imaging speeds, while two superimposed depletion patterns achieve axial and highly parallelized lateral superresolution. It is unclear to which extend amoeboid cells rely on adhesion receptors vs remodeling of actin cytoskeleton during 3D migration. I will label membrane, actin and adhesion-associated proteins of both wildtype and talin knockout leukocytes and image migration in a 3D collagen matrix to study: 1) what are the characteristics of actin flow, 2) do cells form adhesions and 3) what forces are transmitted onto the substrate? The proposed project advances both state-of-the-art imaging and understanding of 3D cell migration substantially. My microscope will enable further study of diverse live systems well beyond current methods, including development, cell biology and neuroscience.

Cancer evolves through the continuous acquisition of DNA mutations. As a result, multiple diverse tumor subclones can arise during the course or tumor progression from benign to malignant carcinoma. These diverse cancer subclones have different traits, such as different levels of therapy resistance or metastatic potential. Intriguingly, cancers can obtain a hypermutator phenotype that leads to accelerated acquisition of DNA mutations. In gastric cancers, a medical burden especially in East Asia, multiple types of mutations are observed, suggestive of pronounced impact of hypermutator phenotypes on the development of these cancers. However, how hypermutator action facilitates the evolutionary trajectory of cancers is not fully understood.
In this HFSP project, we intend to use organoid technology to study the impact of diverse hypermutator phenotypes in gastric cancers. Organoid technology is a state-of-the-art culture technique for human mini-organs in a dish from both normal tissues as well as cancers, and mimics the in vivo scenario to great extent.
To understand the role of hypermutator action in human gastric cancer, three research teams integrate their expertise into a novel experimental pipeline: 1) the establishment and analysis of patient-derived tumor organoids with natural occurring hypermutator phenotypes, in parallel with engineered tumor organoids with introduced hypermutator phenotypes. 2) Monitoring mutational accumulation within these tumor organoids using DNA sequencing at every intermediate step along their progression towards malignant carcinoma and 3) filming the diverging cellular behaviors between different tumor subclones using advanced microscopy. Ultimately, we aim to map the dynamic mutation landscape during the evolutionary trajectory of diverging tumor subclones in comparison to the phenotype of the changing tumor cells.
This project will provide a unique opportunity to obtain a comprehensive understanding on the role of hypermutator phenotypes in cancer evolution. Moreover, we expect that our improved insights on the emergence of genetic subclones in gastric cancer can guide us to understand to other cancer subtypes and will help us to fight against chemo resistance and metastasis of cancer.

Biological carbon fixation is a key step in the global carbon cycle that produces our food and fuels, and regulates the atmosphere’s composition. Approximately 30-40 percent of global carbon fixation occurs in an overlooked algal organelle called the pyrenoid. In this work, I will utilize the single-celled green alga C. reinhardtii to explore how the pyrenoid, a non-membrane bound organelle, is separated from its environment in order to carry out its fundamental function. Specifically, I will apply biophysical methods to experimentally test whether the pyrenoid forms by liquid-liquid phase separation. I will discover the necessary and sufficient molecular components for pyrenoid formation, and will explore their ability to form droplets in-vitro, a hallmark of liquid-liquid phase separation proteins. Next, I will characterize the regulatory mechanisms of the pyrenoid dynamic assembly and disassembly. I will explore the mechanisms governing pyrenoid phase switching by mass-spectrometry PTMs (post-translational modification) mapping. To identify regulators, I will use the novel method of proximity proteomic mapping in living cells by the engineered peroxidase (APEX2). By revealing the underlying mechanism for pyrenoid biogenesis and regulation, I will improve our understanding of its activity, a key step towards a systematic understanding of the global regulation of photosynthesis. Additionally, this work will enhance our understanding of formation and function of non-membrane bound organelles, and will expend this field into the photosynthetic kingdom.

De novo design of proteins enables access to the vast regions of sequence space previously unexplored by nature. These proteins are computationally designed based on physicals principles of protein structure and folding, and are tailor-made with a specific function in mind. As a result, it would be possible to design customized proteins that can give insights into the fundamentals of protein biochemistry as well as tackle practical challenges in medicine and materials sciences. To further push the boundaries of de novo protein design, its concepts will be combined with the world of metal-catalysis. As a result, novel functional metalloproteins will be developed with the protein structure designed to support an active site that is most optimal for high reactivity and selectivity. The design efforts will be directed towards both known biochemical transformations as well as those natively not catalyzed by enzymes. To achieve this goal, quantum chemical methods will be used to develop precise active site models for the transition states of the catalyzed reactions. Rosetta protein design algorithms will then be employed to identify sequences that would best support the atomic arrangements of these theoretical enzyme models. Further rounds of sequence optimization will be performed to ensure high stability and predictability of the active site as well as the entire protein structure. Finally best designs will be expressed and experimentally tested for their activity. In general, the knowledge gained in the process will be beneficial for establishing the general underlying principles behind de novo metalloprotein design.

A large fraction of our genome consists of selfish DNA modules known as transposable elements (TEs) – mobile units that aim to increase in copy number by jumping from one location to the other. Since their discovery, TEs have been involved in the organization, functioning, and evolution of genomes, but their uncontrolled activity is detrimental to the host and must therefore be tightly regulated. This is especially true in the germline, the grounds where host and TE mechanisms compete for maximizing their influence over the genetic information that will be passed to the following generations. A classic example of disruptive TE activity is provided by hybrid dysgenesis in Drosophila, a syndrome that specifically affects germline development and that is triggered by the P-element DNA transposon. Using this textbook model of genomic conflict, we have recently uncovered the existence of a novel mechanism by which an evolutionary conserved small RNA system controls TE expression by regulating chromatin states and alternative splicing. Building on this foundation, the proposal first aims to dissect the emerging and exciting relationship between chromatin states, transcription, and splicing regulation in vivo. In parallel, I will use this system in combination with developmental, genetics, and high-throughput molecular approaches to understand, at the single-cell level, how host-mechanisms access and control genome integrity during catastrophic germline damage. Dissecting such mechanisms will not only provide unique insight into our comprehension of germline development, but will help us understand one of the major forces that shape the evolution of eukaryotic genomes.

Animals move by coordinated muscle activity, which they have to learn during early development. The motor learning period involves spontaneous muscle and neuronal activity, neuromuscular coupling, and experience-dependent network plasticity. Through these processes, the brain ‘discovers’ physical constraints of the body, to generate motion. How does the interplay between neuronal dynamics and body mechanics leads to coordinated motion during development remains unclear. Here, I propose to study the role of mechanical feedback during motor learning of the D. melanogaster embryo. I will explore how physical parameters such as muscle force and elasticity are encoded into neural activity during development, and how non-linear tissue mechanics and spontaneous myogenic contractions enrich neural dynamics and drive the emergence of coordinated motion. To test these ideas, I will assemble a novel light-sheet/light-field microscope, which will enable high-frequency 10-50hz in toto imaging of neuronal and muscle activity in the developing embryo. In addition, I will apply optogenetic tools to directly manipulate muscle contraction activity, and study their effect on the learning dynamics. This comprehensive data set will allow me to develop a physical model of motor learning. The model will combine biological parameters such as spontaneous neural activity and network plasticity, together with physical parameters such as muscle force response, tissue elasticity and substrate friction. This research will reveal novel aspects of motor learning, with implications to developmental biology and medical research.

Leukemia is a paradoxical cancer case, being among the most studied but still most lethal types. Current research focuses on unraveling the cell-intrinsic genome alterations leading to malignant transformation of healthy blood to leukemia initiating cells (LICs), while mostly ignoring their interaction with the bone marrow (BM) microenvironment. This is mainly due to the complexity of the hematopoietic niche itself as well as technical challenges associated with BM imaging without distorting tissue architecture. I hypothesize that distinct BM cell population(s) play a critical role in protecting LICs from cytotoxic effects of current treatments driving minimal residual disease and subsequent deadly relapses. Here, I aim to apply novel quantitative deep-tissue imaging with single-cell resolution to localize LICs and reveal the exact cellular composition of the leukemic BM niche at disease progression, remission and relapse. Identifying the relevant cell populations, I then aim to elucidate the underlying molecular players governing this interaction at the protein level. Using genetic manipulation for both LICs and their niche, I aim to block the protective role of the leukemic niche on disease progression and relapse. To recapitulate key features of human acute myeloid leukemia (AML), a novel mouse model established in the host laboratory allowing non-invasive quantitative monitoring of disease progression in living animals will be utilized. Deciphering the identity of leukemic niche cues protecting cancer stem cells, a currently largely unexplored field of research, could help shaping future therapeutic approaches against rapidly-developing lethal cancers.

Non-alcoholic fatty liver disease (NAFLD) affects 25-40% of the world’s population and is associated with significant morbidity and mortality, yet definitive diagnosis requires invasive liver biopsy, and there are currently no approved pharmacologic agents or validated biomarkers. Further, the pathogenesis of the NAFLD remains unknown, precluding the development of diagnostic and therapeutic tools. Current evidence suggests that pathogenesis involves complex interactions between environmental factors, such as diet and lifestyle, with abnormalities in glucose and lipid homeostasis. The gut microbiota, which is a key mediator of environmental exposures to the host, has been shown to be altered in a mouse model of NAFLD. Gut microbial metabolites play a role as signaling molecules in the setting of metabolic disease and have also recently been shown to regulate host chromatin, ultimately affecting gene expression. Thus, NAFLD-associated changes in the microbiota and its function may play a role in disease development via signaling at the level of cellular receptors and chromatin. Here, I propose to 1) identify NAFLD-associated changes in human gut bacterial community composition and function, 2) develop microbiota-based biomarkers with utility in screening, diagnosis, and prognosis, and 3) elucidate the molecular mechanisms by which altered bacterial taxa and metabolites contribute to disease pathogenesis. Completion of this work will impact clinical management of NAFLD, pave the way for the development of microbiota-based therapies, and yield the first evidence of gut microbial control of global host chromatin states in humans.

In many neurodegenerative diseases, synaptic transmission is perturbed. Optogenetics is a powerful strategy to control neuronal function in health and disease using light-activated ion channels called channelrhodopsins. Although optogenetics could be used to overcome synaptic-transmission defects in many psychiatric and neurodegenerative disorders, several factors hold back this strategy, including the need to open the skull for efficient light delivery and the need to minimize light scattering. To overcome these limitations, I propose to integrate light-generating enzymes and light-activated ion channels into a photon-based neurotransmitter system using presynaptic expression of luciferases and postsynaptic expression of channelrhodopsins, respectively. I will develop microfluidic and novel optogenetic tools to target individual neurons in freely behaving animals. I will demonstrate the feasibility of this approach with the well-characterized sensory circuits in the roundworm Caenorhabditis elegans, a powerful model for neurodegenerative diseases. Taken together, these investigations will empower researchers and ultimately clinicians to replace the “flavor” of chemical neurotransmitters by tuning the color of luciferase-emitted photons and the channels that respond to them. I anticipate that this strategy will unleash the full potential of optogenetics for controlling neuronal function during health and disease.