Procedure so that it can be used with a more diverse set of experimental images and cell lines.cence microscopy to visualize three cell components: the cell membrane, nucleus and microtubules [18]. The original pixel size was 0.05 microns, and the images were downsampled for computational efficiency to 0.2 microns. 2D images of eleven cell lines. The data used here are confocal immunofluorescence microscopy 
images of fixed and interphase cells of eleven different cell lines: A-431, U-2OS, U251MG, RT-4, PC-3, Hep-G2, HeLa, CaCo2, A-549, Hek-293 and MCF-7 from the HPA. They are human cell 25033180 lines widely used in current research. The images were acquired as described previously [1], and the cell lines were obtained from ATCC-LGC Promochem (Boras, Sweden) except that the first two were obtained as described previously [1]. The images are analyzed as 8-bit TIFF images, with two files each obtained using a different emission wavelength of fluorescence from a single image field. These two channel files show the cellular probes/organelles used as references: (i) anti-MedChemExpress 79983-71-4 tubulin antibody as internal control and marker of microtubules and (ii) DAPI for the nucleus. Each of the field images is of size 172861728 for the first three cell lines and 204862048 for the rest of eight, and the pixel size is 0.08 HIV-RT inhibitor 1 microns in the sample plane. The field images were then also downsampled for computational efficiency to 0.2 microns.Computational MethodsCell segmentation for cell size calculation and 3D morphology generation. The field images were segmentedMaterials and Methods Data Acquisition3D image data of HeLa cells. We used 3D images of HeLa cells previously obtained by three color confocal immunofluores-into single cell regions using a seeded watershed method on the tubulin channel with the nuclei in the nuclear channel as seeds. The 2D cell and nuclear boundaries were found by thresholding the single cell regions and the nuclei respectively. These were usedFigure 9. Scatter plot of the estimated total amount of polymerized tubulin versus the area of cytosolic space (sum of pixels) for real cells from eleven cell lines. The correlation coefficient for each cell line is shown in the legend. doi:10.1371/journal.pone.0050292.gComparison of Microtubule Distributionsfor cell size calculation and for 3D morphology generation (see below). Point Spread Function (PSF) estimation. The confocal PSF was generated computationally based on a theoretical model using the SVI PSF calculator for the Zeiss LSM 510 confocal microscope for the first three cell lines and the Leica SP5 for the other eight cell lines (http://www.svi.nl/NyquistCalculator). The pinhole size was set to 1 Airy Unit for the Zeiss and 285.16 nm for the Leica. The numerical aperture was 1.4 and the emissionexcitation 
data used to generate the PSF was for the Alexa555 dye (http://probes.invitrogen.com/handbook/boxes/0442.html). The PSF is used to convolve on the generated raw image of distribution of microtubules to account for the digital blurring from microscopy imaging. Centrosome location detection. The 3D coordinate of the centrosome was estimated by breaking the problem into two parts. First, the XY-coordinate was estimated and then the Z-coordinate. The XY-coordinate was chosen as the pixel with the maximum intensity value in the vicinity of the nucleus after smoothing with an averaging filter of size 25 pixels on the tubulin channel image (as for cell image). For the Z-coordinate, we used linear re.Procedure so that it can be used with a more diverse set of experimental images and cell lines.cence microscopy to visualize three cell components: the cell membrane, nucleus and microtubules [18]. The original pixel size was 0.05 microns, and the images were downsampled for computational efficiency to 0.2 microns. 2D images of eleven cell lines. The data used here are confocal immunofluorescence microscopy images of fixed and interphase cells of eleven different cell lines: A-431, U-2OS, U251MG, RT-4, PC-3, Hep-G2, HeLa, CaCo2, A-549, Hek-293 and MCF-7 from the HPA. They are human cell 25033180 lines widely used in current research. The images were acquired as described previously [1], and the cell lines were obtained from ATCC-LGC Promochem (Boras, Sweden) except that the first two were obtained as described previously [1]. The images are analyzed as 8-bit TIFF images, with two files each obtained using a different emission wavelength of fluorescence from a single image field. These two channel files show the cellular probes/organelles used as references: (i) anti-tubulin antibody as internal control and marker of microtubules and (ii) DAPI for the nucleus. Each of the field images is of size 172861728 for the first three cell lines and 204862048 for the rest of eight, and the pixel size is 0.08 microns in the sample plane. The field images were then also downsampled for computational efficiency to 0.2 microns.Computational MethodsCell segmentation for cell size calculation and 3D morphology generation. The field images were segmentedMaterials and Methods Data Acquisition3D image data of HeLa cells. We used 3D images of HeLa cells previously obtained by three color confocal immunofluores-into single cell regions using a seeded watershed method on the tubulin channel with the nuclei in the nuclear channel as seeds. The 2D cell and nuclear boundaries were found by thresholding the single cell regions and the nuclei respectively. These were usedFigure 9. Scatter plot of the estimated total amount of polymerized tubulin versus the area of cytosolic space (sum of pixels) for real cells from eleven cell lines. The correlation coefficient for each cell line is shown in the legend. doi:10.1371/journal.pone.0050292.gComparison of Microtubule Distributionsfor cell size calculation and for 3D morphology generation (see below). Point Spread Function (PSF) estimation. The confocal PSF was generated computationally based on a theoretical model using the SVI PSF calculator for the Zeiss LSM 510 confocal microscope for the first three cell lines and the Leica SP5 for the other eight cell lines (http://www.svi.nl/NyquistCalculator). The pinhole size was set to 1 Airy Unit for the Zeiss and 285.16 nm for the Leica. The numerical aperture was 1.4 and the emissionexcitation data used to generate the PSF was for the Alexa555 dye (http://probes.invitrogen.com/handbook/boxes/0442.html). The PSF is used to convolve on the generated raw image of distribution of microtubules to account for the digital blurring from microscopy imaging. Centrosome location detection. The 3D coordinate of the centrosome was estimated by breaking the problem into two parts. First, the XY-coordinate was estimated and then the Z-coordinate. The XY-coordinate was chosen as the pixel with the maximum intensity value in the vicinity of the nucleus after smoothing with an averaging filter of size 25 pixels on the tubulin channel image (as for cell image). For the Z-coordinate, we used linear re.
