Colin Ingram
Keynote lecture
Will talk about: Working in the clouds: Creating an e-science collaborative environment for neurophysiology
Professor Colin Ingram is a neurobiologist who obtained his PhD from Cambridge University in the field of neuroendocrinology. In 1986 he moved to Bristol University where he held successive positions as an MRC Training Fellow and Royal Society University Research Fellow, before being appointed as Reader in Neurobiology. His main area of research has been in the neurobiology of stress and depression, including work on the neuropharmacological regulation of serotonergic transmission. In 2000 he took up the Chair of Psychobiology at Newcastle University and in 2004 was appointed the Director of the newly created Institute of Neuroscience. Professor Ingram was responsible for creating the Institute which brings together over 60 faculty staff covering a wide range of basic and clinical neuroscience fields. In directing the Institute he has been involved in a wide range of projects and initiatives to support neuroscience. Since 2006 he has been lead for the EPSRC-funded neuroinformatics project to develop a web-based platform for analysis and storage of neurophysiological data, CARMEN. He is a member of the steering group for the UK Node of INCF and is current Honorary Secretary of the British Neuroscience Association, the UK's principal society serving the neuroscience community.
Understanding the way that neural networks function can only be achieved through detailed analysis of the spatio-temporal relationships between activity patterns of neurones within the network. The increasing sophistication of multielectrode and optical recording techniques has provided the means for us to explore the control and organisation of neural networks in ever greater detail. However, whilst developments in neuroinformatics offer new ways in which the large volumes of data generated from such techniques may be manipulated and analysed, there is a growing need to develop an infrastructure to support sharing and collaborative exploitation of these data. Cloud computing provides exciting opportunities for this exploitation with rapidly increasing availability of data storage and compute resource accessed through the web. However, any practical solution to supporting analysis and sharing of time-series neurophysiological data needs to address a number of major challenges, including interoperability of file formats, data visualisation, and effective metadata descriptions. These challenges are being addressed through a number of international research projects, including CARMEN (Code Analysis, Repository and Modelling for e-Neuroscience; www.carmen.org.uk), an open access web platform developed for handling time-series data using distributed computing technology. CARMEN has created a secure collaborative environment in which data can be shared, either between private groups or publicly, and new analysis algorithms can be deployed without restrictions of issues of software and file compatibility. The on-going implementation of an enactment engine is enabling service applications to be linked into more complex and user-defined workflows. The cloud architecture allows the co-location of data and computation and enabling users to conduct their science through a web browser. Furthermore the growing data repository enables maximum exploitation of data that are often difficult and expensive to produce. CARMEN is one example of the shift from small scale science to large scale data integration which will underpin the next major steps in understanding the brain. Infrastructure built on cloud computing will make this integration feasible and will offer new opportunities to build a global community for neuroscience.
