Michael H. Dickinson - US grants

Dickinson 9723424 Flight in insects is a complex form of locomotion that requires an array of physiological and morphological specializations. The research to be done under this award to the PI will focus on how muscles, skeletons, and brains function together to enable flying animals to perform elaborate aerial maneuvers. The research in this proposal focuses on flies, which although using relatively simple brains, are capable of elaborate flight behavior. Flies are unique among all flying animals in possessing specialized sensory structure called halteres that serve as a gyroscopes during flight. The first goal of the research is to study how the halteres are used to increase flight stability. These experiments will employ a mechanically-oscillating flight arena to characterize the behavioral responses elicited by angular perturbations. The second goal of the project is to use a large scaled model of a flapping fly to determine how changes in wing motion effect the production of aerodynamic forces. The third goal of the research is to combine electrophysiology with high-speed video imaging to characterize how steering muscles act to change the motion of the wings during flight. These first three sets of experiments will determine how animals use corrective reflexes to maintain stable flight. The final portion of the project is to determine how animals initiate voluntary maneuvers when they are actively searching their environment for food or mates. This research will contribute to the understanding of the rapid neural processing and aerodynamic mechanisms that are required to produce sophisticated flight behavior.

Whether winding through alpine meadows, migrating across continents, or circling garbage cans, insects display impressive aerodynamic agility. Although the science of aeronautics is sophisticated enough to design airliners, space shuttles, and stealth fighters, scientists are only just beginning to understand the aerodynamic mechanisms that enable tiny insects to fly and maneuver. This research program builds upon recent discoveries using a variety of experimental and theoretical techniques to construct a comprehensive theory of animal flight. The techniques used in this investigation include three-dimensional high speed videography, with which it is possible to capture the complex wing motions of tiny insects such as fruit flies as they actively steer and maneuver. The research also employs a giant robotic model of flapping insect wings, immersed in a 3 ton tank of mineral oil. By 'replaying' the wing and body motion of real insects on the large robot, the researchers can directly measure the flows and forces created by flapping wings. Through such experiments it will be possible to determine not simply how insects manage to stay in the air, but how they carefully manipulate aerodynamic forces to actively steer and maneuver. Whereas much previous work on insect flight has focused on a small number of species, this research will investigate how the aerodynamic mechanisms used by insects vary with body size, wing shape, and flight speed. Because the physics of air flow can change with scale, this broad comparative analysis is necessary to construct a comprehensive theory of insect flight.
Insects are among the most diverse groups of organisms on the planet, and their flight behavior plays a central role in their extraordinary success. Thus, by forging a clearer picture of how they fly, this research will greatly extend our understanding of this ecologically and agriculturally important group of animals. In addition, just as with complex weather systems, predicting the complex patterns of forces and flows created by flapping wings represents a challenging benchmark for computer simulations in the important and challenging field of Fluid Mechanics - the branch of Physics that determines such diverse phenomena as aerodynamics, heat flow, weather, and global warming. By providing experimental verification of the solutions to complicated flow problems, this research will help mathematicians around the world improve the accuracy of their computer models. Further, knowledge gathered in this study on the aerodynamics of flapping wings will provide new and creative design concepts for the aeronautics industry. Lessons from insect aerodynamics and sensory physiology are already being used in the design of small autonomous air vehicles, whose potential applications include search and rescue operations and planetary exploration. By moving towards a more comprehensive theory of flapping flight aerodynamics, this research will present engineers with a useful body of theory for the development of novel aircraft.

[unreadable] DESCRIPTION (provided by applicant): This project proposes to develop a new class of automated instruments for quantifying complex animal behavior quickly and efficiently, an effort that will be used for many health issues related to human behavior. The effort will focus on the fruit fly, Drosophila melanogaster, because it currently represents the best opportunity for integrating state-of-the-art genetic techniques with a new generation of behavioral assays. Recent research has proven the fruit fly to be a powerful model system for studying many clinically-relevant features of human behavior. This strategy makes use of the fact that the nerve cells of flies and human share many common genetic features, which may be identified quite readily using tools available in fruit flies. For example, investigations that exploit the powerful genetic tools available in this organism have identified a series of candidate genes involved in alcohol and drug tolerance. Recent studies of human populations indicate similar genes may be involved in alcohol addiction. This strategy is not restricted to studies of drug and alcohol addiction, but has also been successfully exploited to study other features of human biology including aging, obesity, fear, aggression and sleep disorders. The potential and impact of these approaches is based on the ease of genetic analyses in flies and also on the ability to accurately identify behavioral defects in large numbers of animals. The goal of this grant will be to thoroughly modernize the quantification of fly behavior so that this genetic model organism can be used more efficiently in the study of a variety of behaviors related to human health. [unreadable] [unreadable] The project team will design, test, and make available three distinct systems that collectively will permit high-throughput quantitative analysis of the individual and social behaviors of adult Drosophila. In designing these instruments, the project team will make use of their collective expertise in machine vision as well as years of practical experience building specialized instruments for quantifying fly behavior. The devices will be intelligent, in that they will be able to automatically identify the most useful measurements to be carried out for a given task, as well as suggest possibly novel models of behavior. They will be based on off-the-shelf video and computer technology to be inexpensive and thus easy to use for the average molecular biologist. Much effort will be made to ensure that the resulting technology will be of use to a broad international community of researchers. Like the robotic sequencers that revolutionized the study of genomics, these devices will help transform behavioral science into a modern discipline of Ethometrics. [unreadable] [unreadable] [unreadable]

One of the great challenges facing modern biology is understanding how the brains of animals coordinate simple motor acts into complex behaviors. For example, the act of a fly leaping from a table and landing on the ceiling requires an intricate sequence of leg and wing motion rapidly modified by sensory feedback. How brains, even those as simple as a fly's, flawlessly execute such feats remains unknown. This project will combine advances in genetics and engineering to experimentally control the activity of individual neurons within the brains of animals that are behaving normally, thus allowing scientists to directly observe the behavioral consequences of specific brain circuits. This multidisciplinary research will involve three components. First, researchers will engineer ion channels within individual neurons that can be opened and closed with pulses of light, creating a switch that turns brain cells on and off. Second, they will genetically engineer animals with these controllable neurons. Finally, they will use sophisticated electronic devices, such as virtual reality flight simulators, that measure changes in behavior resulting from the experimental manipulation of specific brain circuits. This will allow the investigative team to map and decipher brain regions responsible for the control of various behaviors.

The research will be conducted on fruit flies, an important laboratory organism that is used in a wide range of genetic and medical research and is crucial to studying a wide variety of human diseases including alcoholism, senility and obesity. The work will also be incorporated into efforts to 'reverse engineer' flies and combine information about the brains, bodies and behavior of flies to create autonomous flying robots. An educational training and outreach program will foster the development of students and young scientists broadly skilled in biology, math and engineering and capable of approaching formerly intractable problems that require interdisciplinary approaches. This project is a collaboration led by California Institute of Technology (Michael Dickinson) and University of California-Berkeley (Isacoff).

This award is funded under the American Recovery and Reinvestment Act of 2009 (Public Law 111-5).

The ability of an animal to maneuver can determine its success at avoiding predators, catching food, and other fundamental behaviors that define the margin between life and death. Most research on the biomechanics of animal motion has concerned the initiation or maintenance of ballistic, brief, or steady state movements because these can be studied most readily in the laboratory. Maneuverability is therefore one of the most important but least understood aspects of animal locomotion. Previous research has progressed along two independent tracks: 1) Studies of animal morphology have explained how the size and shape of the body and the limbs influence the efficiency and dynamics of maneuvering. 2) Studies of neuromuscular physiology have revealed the mechanisms that animals use to power particular maneuvers. However, animals with the ability to generate substantial muscle power, such as those that hover or fly slowly, may be able to overcome efficiency costs imposed by suboptimal morphology to generate rapid but inefficient maneuvers. The proposed work will test the hypothesis that the limitations imposed by morphology are strongest when muscle power-generating capacity is low. Research will focus on the remarkable maneuvering flight of hummingbirds because these animals inhabit broad elevational ranges, which provide natural experiments for varying muscle power capacity. Experiments with Anna?s hummingbirds (Calypte anna) in California will determine the effects of elevation and the independent influences of mechanical and metabolic constraints on maneuvering performance. Measurements from the diverse Andean hummingbird fauna will allow for determination of how vastly different morphologies influence maneuvering performance across elevations.

A common requirement of diverse disciplines of biology is to have a means of quantitatively describing behavior. This project utilizes a custom-designed, automated analysis of movement to identify the fundamental building blocks of maneuverability. Developing this approach will provide tools that are broadly applicable for studying complex movement in animals. The educational training will foster the scientific development of a postdoctoral scholar, graduate students, and undergraduate students from under-represented groups. These participants will receive integrative training in computational biology and comparative biomechanics. The results of the research will be used to develop a teaching module for use in upper division undergraduate courses. In addition, results produced by the students and the PIs will be presented via scientific conferences, scholarly publications, and public lectures.

DESCRIPTION (provided by applicant): This project addresses the fundamental question: how does the history of sensory experience affect behavioral decision-making? We address this problem in the neuroethological context of mosquito host seeking, in which the presence of thermal and carbon dioxide signals trigger and direct search behavior. This project has several novel aspects. It will quantitatively explore this synergistic stimulus interaction in the context f a natural behavior and develop models for multisensory integration based both on behavioral and neural data. It will address the spatial variations in the statistical structure of olfactory and thermal stimuli through recordings and direct numerical simulations, and examine whether the universal properties of turbulently advected scalar fields can provide sensory evidence for source location and shape neural responses. We will obtain novel neural recordings in response to multiple time-varying inputs. By a developing tethered flight preparation for mosquitoes, we will be able to record neural activity during constrained flight and directly relate sensory neural responses to behavioral outcomes. The results from this project may help in the design of noninvasive mosquito repellents or attractants and so have an impact on disease transmission. The work may also have impact beyond insect physiology in the design of algorithms for novel sensors in the olfactory domain. Additional broader impacts from this project arise from educational and community engagement through interdisciplinary training of undergraduate and graduate students and postdoctoral fellows; active involvement of our research groups in the broader goals of integrated education and research experiences in the areas of Computational Neuroscience and Neural Engineering through new campus-wide initiatives; communication of our results to the community through a wide variety of social media; and participation in outreach activities to teachers and K-12 classrooms.

? DESCRIPTION (provided by applicant): The goal of the project team is to develop a robust, multi-lab research framework, enabled by large scale imaging, which will lead to principled integrative models of ethologically-relevant behaviors that incorporate a detailed knowledge of individual cell classes. The specific neurobiological question that the team will address is how the brain integrates sensory information in order to guide locomotion in a particular direction. Our strategy is to systematically map and functionally characterize the neural circuits that underlie goal-directed locomotion, using the fruit fly, Drosophila, in order to exploit the convergence of powerful genetic, optical, behavioral, and analytical tools that are available in this species. The proposal focuses primarily on refining functional imaging approaches to map the activity of small brain regions and populations of individual neurons in intact, behaving animals while they respond to a controlled panel of sensory stimuli. We have constructed a strategic plan consisting of seven interrelated research modules that create a flow for discovery that starts with functional imaging and ends with the development of integrative models for sensory-guided behavior. The goal of this proposal is to bring all research modules to the requisite level of maturity for future research. To achieve this goal this project will develop robust, quantitative and high throughput methods for: Functional 2-photon imaging using pan-neural drivers. ArcLight imaging using selected driver lines. Functional 2-photon imaging using pan-neural drivers. Circuit analysis of sensory motor pathways. And a plan for an integrative computational model of sensory-guided locomotion.

Like ancient mariners, many flying animals have the ability to navigate long distances across the globe using cues from the earth and sky. One particularly reliable feature of the sky that animals can use as a compass is the pattern by which light from the sun is polarized as it scatters in the atmosphere. This project will study how animals navigate using polarized skylight, by exploiting state-of-the art genetic and physiological approaches that are available in the common fruit fly, Drosophila melanogaster. The results will provide insight into the general question of how circuits in the brain can make accurate calculations that guide locomotion and other motor actions. Simplified, inexpensive versions of the experimental devices used in the proposed studies will be developed for use in education and outreach. The findings will also benefit ongoing efforts to engineer small, energy-efficient autonomous air vehicles that can perform useful missions for civilian and military applications without the need of GPS.

Using this experimentally powerful genetic model organism, it will be possible to determine how specialized cells and circuits in the brain detect and process polarized light and use it as a compass reference for navigation. The project will make use of custom-built flight simulators, in which animals navigate under an artificial sky that may be experimentally manipulated. The research will also employ recently developed techniques of two-photon calcium imaging in freely flying subjects for measuring the activity of cells within relevant neural circuits of the brain as navigation tasks are being performed, a level of analysis not previously achieved. Results from these studies will be disseminated through publications in peer-reviewed journals, and through presentations at scientific conferences.

Based on the number of extant species, insects are arguably the most successful group of animals on earth and the ability to fly is central to their eminence. Although past research has made great progress in understanding the basic aerodynamic mechanisms employed by insects, we know much less about how these animals control the motion of their flapping wings, which is the basis of their remarkable agility. Insects such as flies are endowed with only a dozen or so tiny muscles at the base of the wing with which to control wing motion. This research project will use a combination of genetic and optical techniques to directly observe how insects use their muscles to adjust the flapping pattern of their wings. The investigations will help uncover one of the greatest mysteries about these extraordinarily successful creatures and provide insight for the design and fabrication of miniature mechanical devices such as insect-sized flying robots. In addition, the outreach and training effort of the project will include support of undergraduate research during the academic year.

This proposal will focus on fruit flies (Drosophila melanogaster), which are capable of rapid aerial maneuvers and are amenable to many genetic approaches for recording and manipulating muscles and motor neurons. In particular, by expressing the genetically-encoded, optical calcium sensor GCaMP, the investigators will directly observe the activity of the entire steering muscle system while animals perform visually-elicited flight maneuvers. The functional role of muscles used for different maneuvers will be investigated with high spatial and temporal resolution by combining electrophysiology with high speed videography. The project will also explicitly test the role of specific muscles using optogenetic reagents to activate motor neurons in intact animals. Collectively, these approaches will determine how insects perform flight maneuvers using such a limited set of muscles and expand our knowledge of the general principles of motor control employed by a group of earth's most successful creatures. Results from the studies will be disseminated through publication in peer-reviewed journals and through presentations at scientific meetings.

A Brain Circuit Program for Understanding the Sensorimotor Basis of Behavior Abstract The Project team's long-term goal is to develop a comprehensive theory of animal behavior that explicitly incorporates neural processes operating across hierarchical levels ? from circuits that regulate the action of individual muscles to those that regulate behavioral sequences and decisions. Our innovative approach is guided by the notion that different brain regions are not linked within a single neuroanatomical tier, but rather constitute a series of hierarchically nested feedback loops. The effort is organized into four Research Projects, each focusing on a different processing stage related to: (1) muscle action, (2) motor patterns, (3) motion guidance, and (4) behavioral sequences. Demonstrating our commitment to team interaction, these Research Projects are not organized according to PIs laboratories, but rather each constitutes a collaborative multi- laboratory effort. The collective expertise of our research team spans the entire nervous system - from the sensory periphery to the motor periphery and was chosen to include experts in every experimental technique we require (molecular genetics, electrophysiology, optical imaging, biomechanics, quantitative behavioral analysis, control theory, and dynamic network theory). We will exploit mathematical approaches ? control theory and dynamic network theory in particular ? that are best suited to model feedback and the flow of information through and among different processing stages in the brain. The four complimentary and integrated Research Projects will focus on ethologically relevant natural behaviors, with an emphasis on recording methods that interrogate the functions of genetically identified neurons in intact, behaving animals ? a rigorous standard that is designed to have the broadest impact on systems neuroscience. Our research exploits a single, experimentally tractable model system (Drosophila melanogaster), in which we can easily study the functions of genetically identified cell classes in ethologically relevant behaviors. Our experiments emphasize methods that interrogate the functions of neurons in intact, behaving animals, a rigorous standard that is designed to have the broadest impact on systems neuroscience. Our research will be supported by an Instrumentation and Software Resource Core that will develop and support novel devices and software, so that we can continue to employ state-of-the-art experimental techniques and data analysis. Collectively, our research program constitutes a systematic attack on the neural basis of behavior that integrates vertically across phenomenological tiers. The result of our effort will be a new synthesis of how a fully embodied brain works to generate behavior.  

Core 3: Instrumentation & Software Resource Core Abstract The goal of the Instrumentation and Software Resource Core (ISRC) is to support our team research efforts by developing and maintaining custom experimental instruments and analysis software. Dr. Michael Dickinson will serve as lead PI for the Core, and we will recruit a staff member with training in computer engineering and instrumentation to run the Core. The ISRC will design, fabricate, and support new instruments, including the development of analysis software. Caltech will provide support for the ISRC effort by providing facilities available through their existing Center for Neurotechnology. Although the primary goal the ISRC is to support the research within our team project, our design strategy will ensure that all the tools we develop will be disseminate freely through the research community. The ISRC will work in concert with the Data Sciences Resource Core (DSRT) to ensure that all data formats from new instruments satisfy the requirements for ease of storage and dissemination. Through principled design we will ensure that all new instrumentation employs state-of-the art standards for compression, short and long-term storage capabilities, and format longevity. By developing all instruments using industry-standard CAD software formats and the use of modern fabrication practices, we will make all of our instruments and associated software available to the general neuroscience community. For the most challenging tasks, such as the development of virtual reality systems or real-time machine vision behavioral tracking, we will utilize a powerful set of software tools called ROS (Robot Operating System, www.ros.org), which is a framework for software development specialized for robotic applications.  

Core 1: Administrative Core Abstract The Administrative Core will provide the administrative structure, oversight, and logistical support for the entire Team BCP effort, with the goal of creating synergy among member labs, cores, and projects to best facilitate the proposed research. The Administrative Core includes support for the annual visits of the External Advisory Board to Caltech, travel for the off-site member of the team to attend all Advisory Board Meetings, and maintenance of a website dedicated to the Team BCP. This Core will also coordinate with both the Data Science Resource Core and the Instrumentation and Software Resource Core to ensure equitable distribution of resources to all member laboratories. The Lead PI of this Core, Dr. Michael Dickinson, together with an administrative staff, will provide logistical support for the project. These tasks range from fiscal oversight, to scheduling meetings with member PIs at other institutions, to coordination of efforts with the other project cores. Working with all member PIs, the Administrative Core will be responsible for recruiting an external scientific advisory board and running an annual meeting to be held at Caltech. The Core will arrange travel, accommodations, and logistics. This Core will ensure the dissemination of all products related to the research, including prompt filing of progress reports, ensuring all publications are uploaded to PubMed, and other related administrative and communications functions including maintenance of the program website. In the case of press coverage of any research resulting from the program, we will work with all PIs and the laboratories to ensure that NIH is acknowledged in an appropriate manner.  

Project Summary Although behavior is largely structured by sensory cues, animals also benefit from random actions. Stochastic actions and variability are important for many behaviors including motor learning, search, and predator evasion. Furthermore, understanding the cellular basis of stochasticity may be pertinent to many neurological disorders involving the inability to either suppress or generate involuntary actions, including Tourette's syndrome, Huntington's chorea, and Parkinson's disease. Despite its importance, the mechanisms underlying the generation of random actions at the molecular, cellular, or network levels are poorly understood. Understanding spontaneous processes in the nervous system will benefit from research in a tractable genetic model in which both its functional role and mechanistic basis could be studied vertically across phenomenological levels. Recently, a genetically identified neuron has been identified cell (called DNa04) in the fruit fly, Drosophila that serves as a command neuron for rapid spontaneous turns, called saccades. This is a rare case of a single genetically identified neuron whose activity is necessary and sufficient for stochastic actions. Although it is already possible to record from this command neuron using 2-photon imaging and in vivo patch clamp, the primary goal of this proposal is to identify members of the upstream network that is responsible for generating its stochastic activity. Toward these goals, Specific Aim 1 of this proposal will focus on screening a collection of ~20 selective split-GAL4 lines labeling interneurons in the Lateral Accessory Lobe (LAL), the upstream neuropil region that is thought responsible for generating the spontaneous activity in the DNa04 cell. We will perform 2- photon functional imaging in intact flying flies to record from LAL interneurons and use genetic tracing techniques including trans-Tango to map the connectivity within the network. In Specific Aim 2 we will explore the function of individual cells in generating the pattern of spontaneous activity by manipulating cell physiology using a variety approaches including optogenetic activation and silencing. Specific Aim 3 will focus on psychophysical experiments aimed at quantifying and modeling the influence of sensory stimuli that modulate the frequency of spontaneous saccades. By mapping the network and experimentally manipulating the physiology of its member cells, we will lay the foundation for developing a quantitative model of the cellular basis of spontaneous actions in the brain.