SenseLab: Integration of Multidisciplinary Sensory Data

点击翻译按钮获取中文摘要

SenseLab Renewal Summary The overall aim of SenseLab is to integrate multidisciplinary neuroscience data by means of innovative databases and tools, using the olfactory system as a model which can generalize across the nervous system. For this purpose we have created 8 interoperable databases that serve growing user communities for experimental data and computational models at multiple levels, from genes and proteins through neurons to circuits. SenseLab has three foundations: neuroinformatics directed by Perry Miller, experimental data by Gordon Shepherd, and computational modeling by Michael Hines. One focus will be on ModelDB, which is growing strongly with over 800 computational models. We will build new functionality to enable the models to be explored with new tools including ModelSearch and ModelView. We will support an emerging field of brain microcircuits through MicrocircuitDB, which currently contains over 200 models. A new BrainPathPhysiolDB will contain over 100 models of neuron pathophysiology with clinical relevance. A new ORModelDB will add molecular models to the Olfactory Receptor Database (ORDB) to enhance the utility of the 14,000+ chemosensory genes and proteins that it currently contains. To enhance interoperation we will continue to work closely with the Neuroscience Information Framework (NIF) and the International Neuroinformatics Coordination Facility (INCF) to develop a general ontology for neurons and microcircuits. Support by Dr. Miller and his colleagues in the Yale Center for Medical Informatics will be critical, and enable SenseLab to continue developing its state-of-the-art infrastructure and tools for database construction and interoperation. We will explore innovative ways in which individual SenseLab databases can be designed, adapted, and/or enhanced to facilitate robust interoperation with other neuroscience databases, tools, and resources such as the NIF. In our experimental and computational studies we will develop a new generation of large-scale microcircuit models which realistically represent the detailed 3 dimensional morphology of multiple neuron types with overlapping dendritic fields and distributed synaptic interactions. The NEURON simulator, developed by Dr. Hines, is unique in its capability for computing this model on massively parallel cluster computers. We will test the model with experimental data from an ongoing collaboration with the lab of Dr. Justus Verhagen, and will share the model with other labs working on the olfactory bulb and other systems. In summary, this multidisciplinary and multilevel approach using integration of experimental data into realistic computational simulations should serve as a model for analysis of olfactory processing and for current attempts at data integration throughout the brain.