A new analog trigger system for the Cherenkov Telescope array

La astronomía de rayos γ estudia las partículas más energéticas que llegan a la Tierra desde el espacio. Estos rayos γ no se generan mediante procesos térmicos en simples estrellas, sino mediante mecanismos de aceleración de partículas en objetos celestes como núcleos de galaxias activos, púlsares, supernovas, o posibles procesos de aniquilación de materia oscura. Los rayos γ procedentes de estos objetos y sus características proporcionan una valiosa información con la que los científicos tratan de comprender los procesos físicos que ocurren en ellos y desarrollar modelos teóricos que describan su funcionamiento con fidelidad. El problema de observar rayos γ es que son absorbidos por las capas altas de la atmósfera y no llegan a la superficie (de lo contrario, la Tierra será inhabitable). De este modo, sólo hay dos formas de observar rayos γ embarcar detectores en satélites, u observar los efectos secundarios que los rayos γ producen en la atmósfera. Cuando un rayo γ llega a la atmósfera, interacciona con las partículas del aire y genera un par electrón - positrón, con mucha energía. Estas partículas secundarias generan a su vez más partículas secundarias cada vez menos energéticas. Estas partículas, mientras aún tienen energía suficiente para viajar más rápido que la velocidad de la luz en el aire, producen una radiación luminosa azulada conocida como radiación Cherenkov durante unos pocos nanosegundos. Desde la superficie de la Tierra, algunos telescopios especiales, conocidos como telescopios Cherenkov o IACTs (Imaging Atmospheric Cherenkov Telescopes), son capaces de detectar la radiación Cherenkov e incluso de tomar imágenes de la forma de la cascada Cherenkov. A partir de estas imágenes es posible conocer las principales características del rayo γ original, y con suficientes rayos se pueden deducir características importantes del objeto que los emitió, a cientos de años luz de distancia. Sin embargo, detectar cascadas Cherenkov procedentes de rayos γ no es nada fácil. Las cascadas generadas por fotones γ de bajas energías emiten pocos fotones, y durante pocos nanosegundos, y las correspondientes a rayos γ de alta energía, si bien producen más electrones y duran más, son más improbables conforme mayor es su energía. Esto produce dos líneas de desarrollo de telescopios Cherenkov: Para observar cascadas de bajas energías son necesarios grandes reflectores que recuperen muchos fotones de los pocos que tienen estas cascadas. Por el contrario, las cascadas de altas energías se pueden detectar con telescopios pequeños, pero conviene cubrir con ellos una superficie grande en el suelo para aumentar el número de eventos detectados. Con el objetivo de mejorar la sensibilidad de los telescopios Cherenkov actuales, en el rango de energía alto (> 10 TeV), medio (100 GeV - 10 TeV) y bajo (10 GeV - 100 GeV), nació el proyecto CTA (Cherenkov Telescope Array). Este proyecto en el que participan más de 27 países, pretende construir un observatorio en cada hemisferio, cada uno de los cuales contará con 4 telescopios grandes (LSTs), unos 30 medianos (MSTs) y hasta 70 pequeños (SSTs). Con un array así, se conseguirán dos objetivos. En primer lugar, al aumentar drásticamente el área de colección respecto a los IACTs actuales, se detectarán más rayos γ en todos los rangos de energía. En segundo lugar, cuando una misma cascada Cherenkov es observada por varios telescopios a la vez, es posible analizarla con mucha más precisión gracias a las técnicas estereoscópicas. La presente tesis recoge varios desarrollos técnicos realizados como aportación a los telescopios medianos y grandes de CTA, concretamente al sistema de trigger. Al ser las cascadas Cherenkov tan breves, los sistemas que digitalizan y leen los datos de cada píxel tienen que funcionar a frecuencias muy altas (≈1 GHz), lo que hace inviable que funcionen de forma continua, ya que la cantidad de datos guardada será inmanejable. En su lugar, las señales analógicas se muestrean, guardando las muestras analógicas en un buffer circular de unos pocos µs. Mientras las señales se mantienen en el buffer, el sistema de trigger hace un análisis rápido de las señales recibidas, y decide si la imagen que hay en el buér corresponde a una cascada Cherenkov y merece ser guardada, o por el contrario puede ignorarse permitiendo que el buffer se sobreescriba. La decisión de si la imagen merece ser guardada o no, se basa en que las cascadas Cherenkov producen detecciones de fotones en píxeles cercanos y en tiempos muy próximos, a diferencia de los fotones de NSB (night sky background), que llegan aleatoriamente. Para detectar cascadas grandes es suficiente con comprobar que más de un cierto número de píxeles en una región hayan detectado más de un cierto número de fotones en una ventana de tiempo de algunos nanosegundos. Sin embargo, para detectar cascadas pequeñas es más conveniente tener en cuenta cuántos fotones han sido detectados en cada píxel (técnica conocida como sumtrigger). El sistema de trigger desarrollado en esta tesis pretende optimizar la sensibilidad a bajas energías, por lo que suma analógicamente las señales recibidas en cada píxel en una región de trigger y compara el resultado con un umbral directamente expresable en fotones detectados (fotoelectrones). El sistema diseñado permite utilizar regiones de trigger de tamaño seleccionable entre 14, 21 o 28 píxeles (2, 3, o 4 clusters de 7 píxeles cada uno), y con un alto grado de solapamiento entre ellas. De este modo, cualquier exceso de luz en una región compacta de 14, 21 o 28 píxeles es detectado y genera un pulso de trigger. En la versión más básica del sistema de trigger, este pulso se distribuye por toda la cámara de forma que todos los clusters sean leídos al mismo tiempo, independientemente de su posición en la cámara, a través de un delicado sistema de distribución. De este modo, el sistema de trigger guarda una imagen completa de la cámara cada vez que se supera el número de fotones establecido como umbral en una región de trigger. Sin embargo, esta forma de operar tiene dos inconvenientes principales. En primer lugar, la cascada casi siempre ocupa sólo una pequeña zona de la cámara, por lo que se guardan muchos píxeles sin información alguna. Cuando se tienen muchos telescopios como será el caso de CTA, la cantidad de información inútil almacenada por este motivo puede ser muy considerable. Por otro lado, cada trigger supone guardar unos pocos nanosegundos alrededor del instante de disparo. Sin embargo, en el caso de cascadas grandes la duración de las mismas puede ser bastante mayor, perdiéndose parte de la información debido al truncamiento temporal. Para resolver ambos problemas se ha propuesto un esquema de trigger y lectura basado en dos umbrales. El umbral alto decide si hay un evento en la cámara y, en caso positivo, sólo las regiones de trigger que superan el nivel bajo son leídas, durante un tiempo más largo. De este modo se evita guardar información de píxeles vacíos y las imágenes fijas de las cascadas se pueden convertir en pequeños \vídeos" que representen el desarrollo temporal de la cascada. Este nuevo esquema recibe el nombre de COLIBRI (Concept for an Optimized Local Image Building and Readout Infrastructure), y se ha descrito detalladamente en el capítulo 5. Un problema importante que afecta a los esquemas de sumtrigger como el que se presenta en esta tesis es que para sumar adecuadamente las señales provenientes de cada píxel, estas deben tardar lo mismo en llegar al sumador. Los fotomultiplicadores utilizados en cada píxel introducen diferentes retardos que deben compensarse para realizar las sumas adecuadamente. El efecto de estos retardos ha sido estudiado, y se ha desarrollado un sistema para compensarlos. Por último, el siguiente nivel de los sistemas de trigger para distinguir efectivamente las cascadas Cherenkov del NSB consiste en buscar triggers simultáneos (o en tiempos muy próximos) en telescopios vecinos. Con esta función, junto con otras de interfaz entre sistemas, se ha desarrollado un sistema denominado Trigger Interface Board (TIB). Este sistema consta de un módulo que irá montado en la cámara de cada LST o MST, y que estará conectado mediante fibras ópticas a los telescopios vecinos. Cuando un telescopio tiene un trigger local, este se envía a todos los vecinos conectados y viceversa, de modo que cada telescopio sabe si sus vecinos han dado trigger. Una vez compensadas las diferencias de retardo debidas a la propagación en las fibras ópticas y de los propios fotones Cherenkov en el aire dependiendo de la dirección de apuntamiento, se buscan coincidencias, y en el caso de que la condición de trigger se cumpla, se lee la cámara en cuestión, de forma sincronizada con el trigger local. Aunque todo el sistema de trigger es fruto de la colaboración entre varios grupos, fundamentalmente IFAE, CIEMAT, ICC-UB y UCM en España, con la ayuda de grupos franceses y japoneses, el núcleo de esta tesis son el Level 1 y la Trigger Interface Board, que son los dos sistemas en los que que el autor ha sido el ingeniero principal. Por este motivo, en la presente tesis se ha incluido abundante información técnica relativa a estos sistemas. Existen actualmente importantes líneas de desarrollo futuras relativas tanto al trigger de la cámara (implementación en ASICs), como al trigger entre telescopios (trigger topológico), que darán lugar a interesantes mejoras sobre los diseños actuales durante los próximos años, y que con suerte serán de provecho para toda la comunidad científica participante en CTA. ABSTRACT -ray astronomy studies the most energetic particles arriving to the Earth from outer space. This -rays are not generated by thermal processes in mere stars, but by means of particle acceleration mechanisms in astronomical objects such as active galactic nuclei, pulsars, supernovas or as a result of dark matter annihilation processes. The γ rays coming from these objects and their characteristics provide with valuable information to the scientist which try to understand the underlying physical fundamentals of these objects, as well as to develop theoretical models able to describe them accurately. The problem when observing rays is that they are absorbed in the highest layers of the atmosphere, so they don't reach the Earth surface (otherwise the planet would be uninhabitable). Therefore, there are only two possible ways to observe γ rays: by using detectors on-board of satellites, or by observing their secondary effects in the atmosphere. When a γ ray reaches the atmosphere, it interacts with the particles in the air generating a highly energetic electron-positron pair. These secondary particles generate in turn more particles, with less energy each time. While these particles are still energetic enough to travel faster than the speed of light in the air, they produce a bluish radiation known as Cherenkov light during a few nanoseconds. From the Earth surface, some special telescopes known as Cherenkov telescopes or IACTs (Imaging Atmospheric Cherenkov Telescopes), are able to detect the Cherenkov light and even to take images of the Cherenkov showers. From these images it is possible to know the main parameters of the original -ray, and with some -rays it is possible to deduce important characteristics of the emitting object, hundreds of light-years away. However, detecting Cherenkov showers generated by γ rays is not a simple task. The showers generated by low energy -rays contain few photons and last few nanoseconds, while the ones corresponding to high energy -rays, having more photons and lasting more time, are much more unlikely. This results in two clearly differentiated development lines for IACTs: In order to detect low energy showers, big reflectors are required to collect as much photons as possible from the few ones that these showers have. On the contrary, small telescopes are able to detect high energy showers, but a large area in the ground should be covered to increase the number of detected events. With the aim to improve the sensitivity of current Cherenkov showers in the high (> 10 TeV), medium (100 GeV - 10 TeV) and low (10 GeV - 100 GeV) energy ranges, the CTA (Cherenkov Telescope Array) project was created. This project, with more than 27 participating countries, intends to build an observatory in each hemisphere, each one equipped with 4 large size telescopes (LSTs), around 30 middle size telescopes (MSTs) and up to 70 small size telescopes (SSTs). With such an array, two targets would be achieved. First, the drastic increment in the collection area with respect to current IACTs will lead to detect more -rays in all the energy ranges. Secondly, when a Cherenkov shower is observed by several telescopes at the same time, it is possible to analyze it much more accurately thanks to the stereoscopic techniques. The present thesis gathers several technical developments for the trigger system of the medium and large size telescopes of CTA. As the Cherenkov showers are so short, the digitization and readout systems corresponding to each pixel must work at very high frequencies (_ 1 GHz). This makes unfeasible to read data continuously, because the amount of data would be unmanageable. Instead, the analog signals are sampled, storing the analog samples in a temporal ring buffer able to store up to a few _s. While the signals remain in the buffer, the trigger system performs a fast analysis of the signals and decides if the image in the buffer corresponds to a Cherenkov shower and deserves to be stored, or on the contrary it can be ignored allowing the buffer to be overwritten. The decision of saving the image or not, is based on the fact that Cherenkov showers produce photon detections in close pixels during near times, in contrast to the random arrival of the NSB phtotons. Checking if more than a certain number of pixels in a trigger region have detected more than a certain number of photons during a certain time window is enough to detect large showers. However, taking also into account how many photons have been detected in each pixel (sumtrigger technique) is more convenient to optimize the sensitivity to low energy showers. The developed trigger system presented in this thesis intends to optimize the sensitivity to low energy showers, so it performs the analog addition of the signals received in each pixel in the trigger region and compares the sum with a threshold which can be directly expressed as a number of detected photons (photoelectrons). The trigger system allows to select trigger regions of 14, 21, or 28 pixels (2, 3 or 4 clusters with 7 pixels each), and with extensive overlapping. In this way, every light increment inside a compact region of 14, 21 or 28 pixels is detected, and a trigger pulse is generated. In the most basic version of the trigger system, this pulse is just distributed throughout the camera in such a way that all the clusters are read at the same time, independently from their position in the camera, by means of a complex distribution system. Thus, the readout saves a complete camera image whenever the number of photoelectrons set as threshold is exceeded in a trigger region. However, this way of operating has two important drawbacks. First, the shower usually covers only a little part of the camera, so many pixels without relevant information are stored. When there are many telescopes as will be the case of CTA, the amount of useless stored information can be very high. On the other hand, with every trigger only some nanoseconds of information around the trigger time are stored. In the case of large showers, the duration of the shower can be quite larger, loosing information due to the temporal cut. With the aim to solve both limitations, a trigger and readout scheme based on two thresholds has been proposed. The high threshold decides if there is a relevant event in the camera, and in the positive case, only the trigger regions exceeding the low threshold are read, during a longer time. In this way, the information from empty pixels is not stored and the fixed images of the showers become to little \`videos" containing the temporal development of the shower. This new scheme is named COLIBRI (Concept for an Optimized Local Image Building and Readout Infrastructure), and it has been described in depth in chapter 5. An important problem affecting sumtrigger schemes like the one presented in this thesis is that in order to add the signals from each pixel properly, they must arrive at the same time. The photomultipliers used in each pixel introduce different delays which must be compensated to perform the additions properly. The effect of these delays has been analyzed, and a delay compensation system has been developed. The next trigger level consists of looking for simultaneous (or very near in time) triggers in neighbour telescopes. These function, together with others relating to interfacing different systems, have been developed in a system named Trigger Interface Board (TIB). This system is comprised of one module which will be placed inside the LSTs and MSTs cameras, and which will be connected to the neighbour telescopes through optical fibers. When a telescope receives a local trigger, it is resent to all the connected neighbours and vice-versa, so every telescope knows if its neighbours have been triggered. Once compensated the delay differences due to propagation in the optical fibers and in the air depending on the pointing direction, the TIB looks for coincidences, and in the case that the trigger condition is accomplished, the camera is read a fixed time after the local trigger arrived. Despite all the trigger system is the result of the cooperation of several groups, specially IFAE, Ciemat, ICC-UB and UCM in Spain, with some help from french and japanese groups, the Level 1 and the Trigger Interface Board constitute the core of this thesis, as they have been the two systems designed by the author of the thesis. For this reason, a large amount of technical information about these systems has been included. There are important future development lines regarding both the camera trigger (implementation in ASICS) and the stereo trigger (topological trigger), which will produce interesting improvements for the current designs during the following years, being useful for all the scientific community participating in CTA.

Robust Real-time Vision-based Aircraft Tracking From Unmanned Aerial Vehicles.

Aircraft tracking plays a key and important role in the Sense-and-Avoid system of Unmanned Aerial Vehicles (UAVs). This paper presents a novel robust visual tracking algorithm for UAVs in the midair to track an arbitrary aircraft at real-time frame rates, together with a unique evaluation system. This visual algorithm mainly consists of adaptive discriminative visual tracking method, Multiple-Instance (MI) learning approach, Multiple-Classifier (MC) voting mechanism and Multiple-Resolution (MR) representation strategy, that is called Adaptive M3 tracker, i.e. AM3. In this tracker, the importance of test sample has been integrated to improve the tracking stability, accuracy and real-time performances. The experimental results show that this algorithm is more robust, efficient and accurate against the existing state-of-art trackers, overcoming the problems generated by the challenging situations such as obvious appearance change, variant surrounding illumination, partial aircraft occlusion, blur motion, rapid pose variation and onboard mechanical vibration, low computation capacity and delayed information communication between UAVs and Ground Station (GS). To our best knowledge, this is the first work to present this tracker for solving online learning and tracking freewill aircraft/intruder in the UAVs.

A Ground-Truth Video Dataset for the Development and Evaluation of Vision-based Sense-and-Avoid systems

The importance of vision-based systems for Sense-and-Avoid is increasing nowadays as remotely piloted and autonomous UAVs become part of the non-segregated airspace. The development and evaluation of these systems demand flight scenario images which are expensive and risky to obtain. Currently Augmented Reality techniques allow the compositing of real flight scenario images with 3D aircraft models to produce useful realistic images for system development and benchmarking purposes at a much lower cost and risk. With the techniques presented in this paper, 3D aircraft models are positioned firstly in a simulated 3D scene with controlled illumination and rendering parameters. Realistic simulated images are then obtained using an image processing algorithm which fuses the images obtained from the 3D scene with images from real UAV flights taking into account on board camera vibrations. Since the intruder and camera poses are user-defined, ground truth data is available. These ground truth annotations allow to develop and quantitatively evaluate aircraft detection and tracking algorithms. This paper presents the software developed to create a public dataset of 24 videos together with their annotations and some tracking application results.

Parallel Computer Vision Algorithms for Graphics Processing Units

La evolución de los teléfonos móviles inteligentes, dotados de cámaras digitales, está provocando una creciente demanda de aplicaciones cada vez más complejas que necesitan algoritmos de visión artificial en tiempo real; puesto que el tamaño de las señales de vídeo no hace sino aumentar y en cambio el rendimiento de los procesadores de un solo núcleo se ha estancado, los nuevos algoritmos que se diseñen para visión artificial han de ser paralelos para poder ejecutarse en múltiples procesadores y ser computacionalmente escalables. Una de las clases de procesadores más interesantes en la actualidad se encuentra en las tarjetas gráficas (GPU), que son dispositivos que ofrecen un alto grado de paralelismo, un excelente rendimiento numérico y una creciente versatilidad, lo que los hace interesantes para llevar a cabo computación científica. En esta tesis se exploran dos aplicaciones de visión artificial que revisten una gran complejidad computacional y no pueden ser ejecutadas en tiempo real empleando procesadores tradicionales. En cambio, como se demuestra en esta tesis, la paralelización de las distintas subtareas y su implementación sobre una GPU arrojan los resultados deseados de ejecución con tasas de refresco interactivas. Asimismo, se propone una técnica para la evaluación rápida de funciones de complejidad arbitraria especialmente indicada para su uso en una GPU. En primer lugar se estudia la aplicación de técnicas de síntesis de imágenes virtuales a partir de únicamente dos cámaras lejanas y no paralelas—en contraste con la configuración habitual en TV 3D de cámaras cercanas y paralelas—con información de color y profundidad. Empleando filtros de mediana modificados para la elaboración de un mapa de profundidad virtual y proyecciones inversas, se comprueba que estas técnicas son adecuadas para una libre elección del punto de vista. Además, se demuestra que la codificación de la información de profundidad con respecto a un sistema de referencia global es sumamente perjudicial y debería ser evitada. Por otro lado se propone un sistema de detección de objetos móviles basado en técnicas de estimación de densidad con funciones locales. Este tipo de técnicas es muy adecuada para el modelado de escenas complejas con fondos multimodales, pero ha recibido poco uso debido a su gran complejidad computacional. El sistema propuesto, implementado en tiempo real sobre una GPU, incluye propuestas para la estimación dinámica de los anchos de banda de las funciones locales, actualización selectiva del modelo de fondo, actualización de la posición de las muestras de referencia del modelo de primer plano empleando un filtro de partículas multirregión y selección automática de regiones de interés para reducir el coste computacional. Los resultados, evaluados sobre diversas bases de datos y comparados con otros algoritmos del estado del arte, demuestran la gran versatilidad y calidad de la propuesta. Finalmente se propone un método para la aproximación de funciones arbitrarias empleando funciones continuas lineales a tramos, especialmente indicada para su implementación en una GPU mediante el uso de las unidades de filtraje de texturas, normalmente no utilizadas para cómputo numérico. La propuesta incluye un riguroso análisis matemático del error cometido en la aproximación en función del número de muestras empleadas, así como un método para la obtención de una partición cuasióptima del dominio de la función para minimizar el error. ABSTRACT The evolution of smartphones, all equipped with digital cameras, is driving a growing demand for ever more complex applications that need to rely on real-time computer vision algorithms. However, video signals are only increasing in size, whereas the performance of single-core processors has somewhat stagnated in the past few years. Consequently, new computer vision algorithms will need to be parallel to run on multiple processors and be computationally scalable. One of the most promising classes of processors nowadays can be found in graphics processing units (GPU). These are devices offering a high parallelism degree, excellent numerical performance and increasing versatility, which makes them interesting to run scientific computations. In this thesis, we explore two computer vision applications with a high computational complexity that precludes them from running in real time on traditional uniprocessors. However, we show that by parallelizing subtasks and implementing them on a GPU, both applications attain their goals of running at interactive frame rates. In addition, we propose a technique for fast evaluation of arbitrarily complex functions, specially designed for GPU implementation. First, we explore the application of depth-image–based rendering techniques to the unusual configuration of two convergent, wide baseline cameras, in contrast to the usual configuration used in 3D TV, which are narrow baseline, parallel cameras. By using a backward mapping approach with a depth inpainting scheme based on median filters, we show that these techniques are adequate for free viewpoint video applications. In addition, we show that referring depth information to a global reference system is ill-advised and should be avoided. Then, we propose a background subtraction system based on kernel density estimation techniques. These techniques are very adequate for modelling complex scenes featuring multimodal backgrounds, but have not been so popular due to their huge computational and memory complexity. The proposed system, implemented in real time on a GPU, features novel proposals for dynamic kernel bandwidth estimation for the background model, selective update of the background model, update of the position of reference samples of the foreground model using a multi-region particle filter, and automatic selection of regions of interest to reduce computational cost. The results, evaluated on several databases and compared to other state-of-the-art algorithms, demonstrate the high quality and versatility of our proposal. Finally, we propose a general method for the approximation of arbitrarily complex functions using continuous piecewise linear functions, specially formulated for GPU implementation by leveraging their texture filtering units, normally unused for numerical computation. Our proposal features a rigorous mathematical analysis of the approximation error in function of the number of samples, as well as a method to obtain a suboptimal partition of the domain of the function to minimize approximation error.

A Vision-based Quadrotor Multi-robot Solution for the Indoor Autonomy Challenge of the 2013 International Micro Air Vehicle Competition

This paper presents a completely autonomous solution to participate in the Indoor Challenge of the 2013 International Micro Air Vehicle Competition (IMAV 2013). Our proposal is a multi-robot system with no centralized coordination whose robotic agents share their position estimates. The capability of each agent to navigate avoiding collisions is a consequence of the resulting emergent behavior. Each agent consists of a ground station running an instance of the proposed architecture that communicates over WiFi with an AR Drone 2.0 quadrotor. Visual markers are employed to sense and map obstacles and to improve the pose estimation based on Inertial Measurement Unit (IMU) and ground optical flow data. Based on our architecture, each robotic agent can navigate avoiding obstacles and other members of the multi-robot system. The solution is demonstrated and the achieved navigation performance is evaluated by means of experimental flights. This work also analyzes the capabilities of the presented solution in simulated flights of the IMAV 2013 Indoor Challenge. The performance of the CVG UPM team was awarded with the First Prize in the Indoor Autonomy Challenge of the IMAV 2013 competition.

A dominant form of inherited retinal degeneration caused by a non-photoreceptor cell-specific mutation

We have isolated a dominant mutation, night blindness a (nba), that causes a slow retinal degeneration in zebrafish. Heterozygous nba fish have normal vision through 2–3 months of age but subsequently become night blind. By 9.5 months of age, visual sensitivity of affected fish may be decreased more than two log units, or 100-fold, as measured behaviorally. Electroretinographic (ERG) thresholds of mutant fish are also raised significantly, and the ERG b-wave shows a delayed implicit time. These defects are due primarily to a late-onset photoreceptor cell degeneration involving initially the rods but eventually the cones as well. Homozygous nba fish display an early-onset neuronal degeneration throughout the retina and elsewhere in the central nervous system. As a result, animals develop with small eyes and die by 4–5 days postfertilization (pf). These latter data indicate that the mutation affecting nba fish is not in a photoreceptor cell-specific gene.

Cannabinoid CB1 receptors and ligands in vertebrate retina: Localization and function of an endogenous signaling system

CB1, a cannabinoid receptor enriched in neuronal tissue, was found in high concentration in retinas of rhesus monkey, mouse, rat, chick, goldfish, and tiger salamander by using a subtype-specific polyclonal antibody. Immunolabeling was detected in the two synaptic layers of the retina, the inner and outer plexiform layers, of all six species examined. In the outer plexiform layer, CB1 was located in and/or on cone pedicles and rod spherules. Labeling was detected in some amacrine cells of all species and in the ganglion cells and ganglion cell axons of all species except fish. In addition, sparse labeling was found in the inner and/or outer segments of the photoreceptors of monkey, mouse, rat, and chick. Using GC/MS to detect possible endogenous cannabinoids, we found 3 nmol of 2-arachidonylglycerol per g of tissue, but no anandamide was detectable. Cannabinoid receptor agonists induced a dramatic reduction in the amplitude of voltage-gated L-type calcium channel currents in identified retinal bipolar cells. The presence and distribution of the CB1 receptor, the large amounts of 2-arachidonylglycerol found, and the effects of cannabinoids on calcium channel activity in bipolar cells suggest a substantive role for an endogenous cannabinoid signaling system in retinal physiology, and perhaps vision in general.

Origins and antiquity of X-linked triallelic color vision systems in New World monkeys

It is known that the squirrel monkey, marmoset, and other related New World (NW) monkeys possess three high-frequency alleles at the single X-linked photopigment locus, and that the spectral sensitivity peaks of these alleles are within those delimited by the human red and green pigment genes. The three alleles in the squirrel monkey and marmoset have been sequenced previously. In this study, the three alleles were found and sequenced in the saki monkey, capuchin, and tamarin. Although the capuchin and tamarin belong to the same family as the squirrel monkey and marmoset, the saki monkey belongs to a different family and is one of the species that is most divergent from the squirrel monkey and marmoset, suggesting the presence of the triallelic system in many NW monkeys. The nucleotide sequences of these alleles from the five species studied indicate that gene conversion occurs frequently and has partially or completely homogenized intronic and exonic regions of the alleles in each species, making it appear that a triallelic system arose independently in each of the five species studied. Nevertheless, a detailed analysis suggests that the triallelic system arose only once in the NW monkey lineage, from a middle wavelength (green) opsin gene, and that the amino acid differences at functionally critical sites among alleles have been maintained by natural selection in NW monkeys for >20 million years. Moreover, the two X-linked opsin genes of howler monkeys (a NW monkey genus) were evidently derived from the incorporation of a middle (green) and a long wavelength (red) allele into one chromosome; these two genes together with the (autosomal) blue opsin gene would immediately enable even a male monkey to have trichromatic vision.

A neural system for human visual working memory

Working memory is the process of actively maintaining a representation of information for a brief period of time so that it is available for use. In monkeys, visual working memory involves the concerted activity of a distributed neural system, including posterior areas in visual cortex and anterior areas in prefrontal cortex. Within visual cortex, ventral stream areas are selectively involved in object vision, whereas dorsal stream areas are selectively involved in spatial vision. This domain specificity appears to extend forward into prefrontal cortex, with ventrolateral areas involved mainly in working memory for objects and dorsolateral areas involved mainly in working memory for spatial locations. The organization of this distributed neural system for working memory in monkeys appears to be conserved in humans, though some differences between the two species exist. In humans, as compared with monkeys, areas specialized for object vision in the ventral stream have a more inferior location in temporal cortex, whereas areas specialized for spatial vision in the dorsal stream have a more superior location in parietal cortex. Displacement of both sets of visual areas away from the posterior perisylvian cortex may be related to the emergence of language over the course of brain evolution. Whereas areas specialized for object working memory in humans and monkeys are similarly located in ventrolateral prefrontal cortex, those specialized for spatial working memory occupy a more superior and posterior location within dorsal prefrontal cortex in humans than in monkeys. As in posterior cortex, this displacement in frontal cortex also may be related to the emergence of new areas to serve distinctively human cognitive abilities.

Cloning of an arylalkylamine N-acetyltransferase (aaNAT1) from Drosophila melanogaster expressed in the nervous system and the gut.

In insects, neurotransmitter catabolism, melatonin precursor formation, and sclerotization involve arylalkylamine N-acetyltransferase (aaNAT, EC 2.3.1.87) activity. It is not known if one or multiple aaNAT enzymes are responsible for these activities. We recently have purified an aaNAT from Drosophila melanogaster. Here, we report the cloning of the corresponding aaNAT cDNA (aaNAT1) that upon COS cell expression acetylates dopamine, tryptamine, and the immediate melatonin precursor serotonin. aaNAT1 represents a novel gene family unrelated to known acetyl-transferases, except in two weakly conserved amino acid motifs. In situ hybridization studies of aaNAT1 mRNA in embryos reveal hybridization signals in the brain, the ventral cord, the gut, and probably in oenocytes, indicating a broad tissue distribution of aaNAT1 transcripts. Moreover, in day/ night studies we demonstrate a diurnal rhythm of melatonin concentration without a clear-cut change in aaNAT1 mRNA levels. The data suggest that tissue-specific regulation of aaNAT1 may be associated with different enzymatic functions and do not exclude the possibility of additional aaNAT genes.

Faithful representation of similarities among three-dimensional shapes in human vision.

Efficient and reliable classification of visual stimuli requires that their representations reside a low-dimensional and, therefore, computationally manageable feature space. We investigated the ability of the human visual system to derive such representations from the sensory input-a highly nontrivial task, given the million or so dimensions of the visual signal at its entry point to the cortex. In a series of experiments, subjects were presented with sets of parametrically defined shapes; the points in the common high-dimensional parameter space corresponding to the individual shapes formed regular planar (two-dimensional) patterns such as a triangle, a square, etc. We then used multidimensional scaling to arrange the shapes in planar configurations, dictated by their experimentally determined perceived similarities. The resulting configurations closely resembled the original arrangements of the stimuli in the parameter space. This achievement of the human visual system was replicated by a computational model derived from a theory of object representation in the brain, according to which similarities between objects, and not the geometry of each object, need to be faithfully represented.

How photons start vision.

Recent studies have elucidated how the absorption of a photon in a rod or cone cell leads to the generation of the amplified neural signal that is transmitted to higher-order visual neurons. Photoexcited visual pigment activates the GTP-binding protein transducin, which in turn stimulates cGMP phosphodiesterase. This enzyme hydrolyzes cGMP, allowing cGMP-gated cationic channels in the surface membrane to close, hyperpolarize the cell, and modulate transmitter release at the synaptic terminal. The kinetics of reactions in the cGMP cascade limit the temporal resolution of the visual system as a whole, while statistical fluctuations in the reactions limit the reliability of detection of dim light. Much interest now focuses on the processes that terminate the light response and dynamically regulate amplification in the cascade, causing the single photon response to be reproducible and allowing the cell to adapt in background light. A light-induced fall in the internal free Ca2+ concentration coordinates negative feedback control of amplification. The fall in Ca2+ stimulates resynthesis of cGMP, antagonizes rhodopsin's catalytic activity, and increases the affinity of the light-regulated cationic channel for cGMP. We are using physiological methods to study the molecular mechanisms that terminate the flash response and mediate adaptation. One approach is to observe transduction in truncated, dialyzed photoreceptor cells whose internal Ca2+ and nucleotide concentrations are under experimental control and to which exogenous proteins can be added. Another approach is to observe transduction in transgenic mouse rods in which specific proteins within the cascade are altered or deleted.

The biology of vision of Drosophila.

Phototransduction systems in vertebrates and invertebrates share a great deal of similarity in overall strategy but differ significantly in the underlying molecular machinery. Both are rhodopsin-based G protein-coupled signaling cascades displaying exquisite sensitivity and broad dynamic range. However, light activation of vertebrate photoreceptors leads to activation of a cGMP-phosphodiesterase effector and the generation of a hyperpolarizing response. In contrast, activation of invertebrate photoreceptors, like Drosophila, leads to stimulation of phospholipase C and the generation of a depolarizing receptor potential. The comparative study of these two systems of phototransduction offers the opportunity to understand how similar biological problems may be solved by different molecular mechanisms of signal transduction. The study of this process in Drosophila, a system ideally suited to genetic and molecular manipulation, allows us to dissect the function and regulation of such a complex signaling cascade in its normal cellular environment. In this manuscript I review some of our recent findings and the strategies used to dissect this process.

Emergence of order in visual system development.

Neural connections in the adult central nervous system are highly precise. In the visual system, retinal ganglion cells send their axons to target neurons in the lateral geniculate nucleus (LGN) in such a way that axons originating from the two eyes terminate in adjacent but nonoverlapping eye-specific layers. During development, however, inputs from the two eyes are intermixed, and the adult pattern emerges gradually as axons from the two eyes sort out to form the layers. Experiments indicate that the sorting-out process, even though it occurs in utero in higher mammals and always before vision, requires retinal ganglion cell signaling; blocking retinal ganglion cell action potentials with tetrodotoxin prevents the formation of the layers. These action potentials are endogenously generated by the ganglion cells, which fire spontaneously and synchronously with each other, generating "waves" of activity that travel across the retina. Calcium imaging of the retina shows that the ganglion cells undergo correlated calcium bursting to generate the waves and that amacrine cells also participate in the correlated activity patterns. Physiological recordings from LGN neurons in vitro indicate that the quasiperiodic activity generated by the retinal ganglion cells is transmitted across the synapse between ganglion cells to drive target LGN neurons. These observations suggest that (i) a neural circuit within the immature retina is responsible for generating specific spatiotemporal patterns of neural activity; (ii) spontaneous activity generated in the retina is propagated across central synapses; and (iii) even before the photoreceptors are present, nerve cell function is essential for correct wiring of the visual system during early development. Since spontaneously generated activity is known to be present elsewhere in the developing CNS, this process of activity-dependent wiring could be used throughout the nervous system to help refine early sets of neural connections into their highly precise adult patterns.