IMAGE DECONVOLUTION The preferred method of live cell imaging in this lab is widefield microscopy using highly attenuated excitation light and  an electron-multiplied charge-coupled device (EM-CCD) camera. This results in optics that see all the potential light signal in a sample, both out of focus, and in focus. the camera is the most light sensitive chip available, and light is amplified prior to hitting the chip. The reason why we use this method is that it requires over 10X less protein expression than confocal methods, which is a major concern in the study of aggregate-prone proteins like mutant huntingtin. Confocal imaging achieves aesthetically pleasing sharp images by discarding 90% or more of the total light signal, thus requiring a considerable amount of protein over-expression. This caveat is rarely considered in cell biological studies of HD. For example, transfection of 1ug of CMV-promoted huntingtin exon1 fragment for 24 hours in NIH3T3 fibroblast cell line will produce ~1,600X molar excess of this fragment over endogenous huntingtin. Many typical exon1 cell biological studies published greatly exceed this amount. Image deconvolution, or image restoration, uses a method in which the path of light through a focal point is understood mathematically through a model called the point spread function (PSF). Light is analyzed in a 3D stack from out of focus to in-focus, and the PSF is applied. By Fourier transform, light intensity is corrected form the out-of-focus plane to the in-focus plane. The result is then analyzed, and the PSF is iteratively corrected, then the process is repeated until a high level of statistical confidence is achieved. The net result is a sharp image, with increased signal-to-noise ratio that can be quantified, and intensity levels equal to confocal imaging with much less protein expressed. This can all be done in multiple channels, and through time for 4D imaging. EM-CCD cameras with large pixel arrays (1024X1024) are tremendously expensive and only recently made available for non-military based research, and require customized optical configurations. We are one of the first labs to successfully develop this system, merging optical technology with computer software seamlessly, with software than can understand a common file format with meta data. Scope 3D, 4D acquisition software: micro-Manager Deconvolution software: Autodeblur X2 3D, 4D, visualization and quantification software: Bitplane Imaris, Image J. Statistical software: Sigmaplot 11, Microsoft Excel, Matlab