2021 article
Automated tracking of S. pombe spindle elongation dynamics
Uzsoy, A. S. M., Zareiesfandabadi, P., Jennings, J., Kemper, A. F., & Elting, M. W. (2021, July 8). JOURNAL OF MICROSCOPY.
author keywords: automation; cell division; fission yeast; fluorescence imaging; image analysis; live-cell imaging; mitotic spindle
MeSH headings : Microscopy; Microtubules / chemistry; Mitosis; Schizosaccharomyces; Spindle Apparatus
doi.org (repository)
Source: Web Of Science
Added: August 16, 2021
The mitotic spindle is a microtubule-based machine that pulls the two identical sets of chromosomes to opposite ends of the cell during cell division. The fission yeast Schizosaccharomyces pombe is an important model organism for studying mitosis due to its simple, stereotyped spindle structure and well-established genetic toolset. S. pombe spindle length is a useful metric for mitotic progression, but manually tracking spindle ends in each frame to measure spindle length over time is laborious and can limit experimental throughput. We have developed an ImageJ plugin that can automatically track S. pombe spindle length over time and replace manual or semi-automated tracking of spindle elongation dynamics. Using an algorithm that detects the principal axis of the spindle and then finds its ends, we reliably track the length of the spindle as the cell divides. The plugin integrates with existing ImageJ features, exports its data for further analysis outside of ImageJ and does not require any programming by the user. Thus, the plugin provides an accessible tool for quantification of S. pombe spindle length that will allow automatic analysis of large microscopy data sets and facilitate screening for effects of cell biological perturbations on mitotic progression. The mitotic spindle is a biological machine that pulls the two identical sets of DNA to opposite ends of the cell during cell division. Incorrect cell division can result in serious issues like cancer and miscarriages. Schizosaccharomyces pombe (S. pombe), a kind of yeast, is commonly used to study cell division because its mitotic spindle is essentially linear in shape and its DNA sequence is well known, allowing for more complex experiments. To measure how well a cell divides, we measure the length of the spindle over time, but this can be tedious to do by hand for many cell images. We have developed software that interfaces with ImageJ (a common image analysis tool) that automatically tracks the length of S. pombe spindles over time and can replace manual tracking. Our software calculates the spindle's lines of symmetry, while allows us to accurately measure the length and track the ends over time. It integrates with existing ImageJ features, exports its data for further analysis outside of ImageJ, and does not require any programming by the user. Thus, the plugin provides an accessible tool for measuring S. pombe spindle length that will allow automatic analysis of large microscopy data sets and facilitate screening for effects of defects in cell division. This will facilitate the study of the basic fundamental process of how cells divide, and could have significant long term medical impacts. Figure S1: Additional results of simulated data analysis for the “long” spindle (190 pixels). (a) Length vs. SNR for a long spindle at an angle of 90 degrees from the x axis. (b) Length vs. orientation angle for a long spindle, at a constant SNR of 15.5. (c) Length vs. orientation angle for a long spindle, at a constant SNR of 2.9. (d) Length vs. orientation angle for a long spindle, at a constant SNR of 1.7. All images are 80 x 247. The blue line indicates the actual spindle length. There were 15 trials for each experimental condition Figure S2: Additional results of simulated data analysis for the “medium” spindle (100 pixels). (a) Length vs. SNR for a medium spindle at an angle of 60 degrees from the x axis. (b) Length vs. SNR for a medium spindle at an angle of 120 degrees from the x axis. (c) Length vs. orientation angle for a medium spindle, at a constant SNR of 15.5. (d) Length vs. orientation angle for a medium spindle, at a constant SNR of 6.8. All images are 80 x 247. The blue line indicates the actual spindle length. There were 15 trials for each experimental condition Figure S3: Additional results of simulated data analysis for the “short” spindle (50 pixels). (a) Length vs. SNR for a short spindle at an angle of 90 degrees from the x axis. (b) Length vs. SNR for a short spindle at an angle of 0 degrees from the x axis. (c) Length vs. orientation angle for a short spindle, at a constant SNR of 15.5 (d) Length vs. orientation angle for a short spindle, at a constant SNR of 6.8. All images are 80 x 247. The blue line indicates the actual spindle length. There were 15 trials for each experimental condition Figure S4: Effect of exposure time and spindle length on SNR. Images of both short (a) and long (b) spindles are shown at exposures times ranging from 50 ms to 800 ms. All images were collected at 7% laser power. SNR, calculated as described in Eq. 1, is shown for each image. For display in the figure, we automatically normalized brightness and contrast for each individual image, so that the dimmest pixel in the image is black and the brightest is white Figure S5: Effect of laser power and spindle length on SNR. Images of both short (a) and long (b) spindles are shown at laser powers ranging from 2% to 40 %. All images were collected with 100 ms exposures. SNR, calculated as described in Eq. 1, is shown for each image. For display in the figure, we automatically normalized brightness and contrast for each individual image, so that the dimmest pixel in the image is black and the brightest is white Data S2 Data S3 Data S4 Data S5 Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
