Stress Granule Quantification
Locked
 
|
Hello,
First, thanks to everyone who contributes here. I have subscribed for several months and learned a ton. ImageJ / FIJI and this list is an absolutely fantastic resource for image processing newbies like myself. I am trying to quantify stress granules in cultured cells by immunofluorescence. My images (links below) are of cells stained for a granule marker. I think the granules are very obvious by eye, but being a good (lazy) scientist I would like to quantitate them in an unbiased (automated) way. My original plan to analyze the images was: -subtract the background using a rolling ball -then use auto local threshold -analyze particles Unfortunately, this doesn't work very well, even after adjusting the many options available to me, including smoothing, size cutoffs, and a few other options I tried but can't remember. I get a high number of detected particles in the negative control. Clearly I need to change my approach to thresholding the image. Any suggestions for different ways to segment and count these granules? No Stress Granules: https://drive.google.com/file/d/0B2o8fWkzLLNQSlVNS0dTTzBxVEk/edit?usp=sharin g Stress Granules: https://drive.google.com/file/d/0B2o8fWkzLLNQNnpFN0VBYVF1QlU/edit?usp=sharin g Note that google drive image preview has horribly borked the images - download and open in FIJI and they look fine. Thanks for any help, Allan Heibein -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
Re: Stress Granule Quantification
|
|
Hello:
In my excitement to post a message to this forum I mislabeled the images. No Stress Granules should be Stress Granules and vice-versa. Thanks again, Allan Heibein -----Original Message----- From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of Allan Heibein Sent: March-11-14 12:12 PM To: [hidden email] Subject: Stress Granule Quantification Hello, First, thanks to everyone who contributes here. I have subscribed for several months and learned a ton. ImageJ / FIJI and this list is an absolutely fantastic resource for image processing newbies like myself. I am trying to quantify stress granules in cultured cells by immunofluorescence. My images (links below) are of cells stained for a granule marker. I think the granules are very obvious by eye, but being a good (lazy) scientist I would like to quantitate them in an unbiased (automated) way. My original plan to analyze the images was: -subtract the background using a rolling ball -then use auto local threshold -analyze particles Unfortunately, this doesn't work very well, even after adjusting the many options available to me, including smoothing, size cutoffs, and a few other options I tried but can't remember. I get a high number of detected particles in the negative control. Clearly I need to change my approach to thresholding the image. Any suggestions for different ways to segment and count these granules? No Stress Granules: https://drive.google.com/file/d/0B2o8fWkzLLNQSlVNS0dTTzBxVEk/edit?usp=sharin g Stress Granules: https://drive.google.com/file/d/0B2o8fWkzLLNQNnpFN0VBYVF1QlU/edit?usp=sharin g Note that google drive image preview has horribly borked the images - download and open in FIJI and they look fine. Thanks for any help, Allan Heibein -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
Re: Stress Granule Quantification
|
|
In reply to this post by Allan Heibein
Hi Allan!
We have developed an algorithm for detecting subcellular structures in fluorescence microscopy images based on Wavelet analysis. We also used it successfully for detecting stress granules and P-bodies. I applied our algorithm to your images, you can find prototypical results here: http://www2.informatik.uni-halle.de/agprbio/StressGranuleDetectExample.tif http://www2.informatik.uni-halle.de/agprbio/NoStressGranuleDetectExample.tif Indeed the algorithm also detects several particles in your non-stress granule image. For now I did not do much tuning on the parameters of the algorithm and I neither tried to apply any post-processing, but maybe in doing so the number of false positives can further be reduced. The algorithm, the ParticleDetectorUWT2D, is available for ImageJ and Fiji as part of MiToBo. You can find download and installation instructions here: http://www2.informatik.uni-halle.de/agprbio/mitobo/index.php/Downloads If you are also interested in the theoretical background you can find the paper here: O. Greß, B. Möller, N. Stöhr, S. Hüttelmaier and S. Posch, "Scale-adaptive Wavelet-based Particle Detection in Microscopy Images". In Proc. of Workshop Bildverarbeitung für die Medizin (BVM '10), Springer, Informatik Aktuell, pp. 266-270, Aachen, Germany, March 2010. http://ceur-ws.org/Vol-574/bvm2010_54.pdf If you require additional support or have further questions or comments, just write an email to us. Cheers, Birgit ------------------------------------------------------------------------ Dr. rer. nat. Birgit Moeller Pattern Recognition & Bioinformatics Institute of Computer Science Faculty of Natural Sciences III Martin Luther University Halle-Wittenberg ------------------------------------------------------------------------ On Tue, 11 Mar 2014 13:59:58 -0400, Allan Heibein <[hidden email]> wrote: >Hello: > >In my excitement to post a message to this forum I mislabeled the images. >No Stress Granules should be Stress Granules and vice-versa. > >Thanks again, > >Allan Heibein > >-----Original Message----- >From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of Allan >Heibein >Sent: March-11-14 12:12 PM >To: [hidden email] >Subject: Stress Granule Quantification > >Hello, > > > >First, thanks to everyone who contributes here. I have subscribed for >several months and learned a ton. ImageJ / FIJI and this list is an >absolutely fantastic resource for image processing newbies like myself. > > > >I am trying to quantify stress granules in cultured cells by >immunofluorescence. My images (links below) are of cells stained for a >granule marker. I think the granules are very obvious by eye, but being a >good (lazy) scientist I would like to quantitate them in an unbiased >(automated) way. My original plan to analyze the images was: > >-subtract the background using a rolling ball > >-then use auto local threshold > >-analyze particles > > > >Unfortunately, this doesn't work very well, even after adjusting the many >options available to me, including smoothing, size cutoffs, and a few other >options I tried but can't remember. I get a high number of detected >particles in the negative control. Clearly I need to change my approach to >thresholding the image. Any suggestions for different ways to segment and >count these granules? > > > >No Stress Granules: > >https://drive.google.com/file/d/0B2o8fWkzLLNQSlVNS0dTTzBxVEk/edit?usp=sharin >g > > > >Stress Granules: > >https://drive.google.com/file/d/0B2o8fWkzLLNQNnpFN0VBYVF1QlU/edit?usp=sharin >g > > > >Note that google drive image preview has horribly borked the images - >download and open in FIJI and they look fine. > > > > > >Thanks for any help, > > > >Allan Heibein > > >-- >ImageJ mailing list: http://imagej.nih.gov/ij/list.html > >-- >ImageJ mailing list: http://imagej.nih.gov/ij/list.html -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
Re: Stress Granule Quantification
|
|
Dear Birgit:
Thank you for your suggestion! Your examples look very encouraging. I will download your plugin and give this a try. Thanks, Allan -----Original Message----- From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of Birgit M öller Sent: March-12-14 5:16 AM To: [hidden email] Subject: Re: Stress Granule Quantification Hi Allan! We have developed an algorithm for detecting subcellular structures in fluorescence microscopy images based on Wavelet analysis. We also used it successfully for detecting stress granules and P-bodies. I applied our algorithm to your images, you can find prototypical results here: <http://www2.informatik.uni-halle.de/agprbio/StressGranuleDetectExample.tif> http://www2.informatik.uni-halle.de/agprbio/StressGranuleDetectExample.tif <http://www2.informatik.uni-halle.de/agprbio/NoStressGranuleDetectExample.tif> http://www2.informatik.uni-halle.de/agprbio/NoStressGranuleDetectExample.tif Indeed the algorithm also detects several particles in your non-stress granule image. For now I did not do much tuning on the parameters of the algorithm and I neither tried to apply any post-processing, but maybe in doing so the number of false positives can further be reduced. The algorithm, the ParticleDetectorUWT2D, is available for ImageJ and Fiji as part of MiToBo. You can find download and installation instructions here: <http://www2.informatik.uni-halle.de/agprbio/mitobo/index.php/Downloads> http://www2.informatik.uni-halle.de/agprbio/mitobo/index.php/Downloads If you are also interested in the theoretical background you can find the paper here: O. Greß, B. Möller, N. Stöhr, S. Hüttelmaier and S. Posch, "Scale-adaptive Wavelet-based Particle Detection in Microscopy Images". In Proc. of Workshop Bildverarbeitung für die Medizin (BVM '10), Springer, Informatik Aktuell, pp. 266-270, Aachen, Germany, March 2010. <http://ceur-ws.org/Vol-574/bvm2010_54.pdf> http://ceur-ws.org/Vol-574/bvm2010_54.pdf If you require additional support or have further questions or comments, just write an email to us. Cheers, Birgit ------------------------------------------------------------------------ Dr. rer. nat. Birgit Moeller Pattern Recognition & Bioinformatics Institute of Computer Science Faculty of Natural Sciences III Martin Luther University Halle-Wittenberg ------------------------------------------------------------------------ On Tue, 11 Mar 2014 13:59:58 -0400, Allan Heibein < <mailto:[hidden email]> [hidden email]> wrote: >Hello: > >In my excitement to post a message to this forum I mislabeled the images. >No Stress Granules should be Stress Granules and vice-versa. > >Thanks again, > >Allan Heibein > >-----Original Message----- >From: ImageJ Interest Group [ <mailto:[hidden email]> mailto:[hidden email]] On Behalf Of >Allan Heibein >Sent: March-11-14 12:12 PM >To: <mailto:[hidden email]> [hidden email] >Subject: Stress Granule Quantification > >Hello, > > > >First, thanks to everyone who contributes here. I have subscribed for >several months and learned a ton. ImageJ / FIJI and this list is an >absolutely fantastic resource for image processing newbies like myself. > > > >I am trying to quantify stress granules in cultured cells by >immunofluorescence. My images (links below) are of cells stained for a >granule marker. I think the granules are very obvious by eye, but being >a good (lazy) scientist I would like to quantitate them in an unbiased >(automated) way. My original plan to analyze the images was: > >-subtract the background using a rolling ball > >-then use auto local threshold > >-analyze particles > > > >Unfortunately, this doesn't work very well, even after adjusting the >many options available to me, including smoothing, size cutoffs, and a >few other options I tried but can't remember. I get a high number of >detected particles in the negative control. Clearly I need to change my >approach to thresholding the image. Any suggestions for different ways >to segment and count these granules? > > > >No Stress Granules: > > <https://drive.google.com/file/d/0B2o8fWkzLLNQSlVNS0dTTzBxVEk/edit?usp=s> https://drive.google.com/file/d/0B2o8fWkzLLNQSlVNS0dTTzBxVEk/edit?usp=s >harin >g > > > >Stress Granules: > > <https://drive.google.com/file/d/0B2o8fWkzLLNQNnpFN0VBYVF1QlU/edit?usp=s> https://drive.google.com/file/d/0B2o8fWkzLLNQNnpFN0VBYVF1QlU/edit?usp=s >harin >g > > > >Note that google drive image preview has horribly borked the images - >download and open in FIJI and they look fine. > > > > > >Thanks for any help, > > > >Allan Heibein > > >-- >ImageJ mailing list: <http://imagej.nih.gov/ij/list.html> http://imagej.nih.gov/ij/list.html > >-- >ImageJ mailing list: <http://imagej.nih.gov/ij/list.html> http://imagej.nih.gov/ij/list.html -- ImageJ mailing list: <http://imagej.nih.gov/ij/list.html> http://imagej.nih.gov/ij/list.html -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
|
|
In reply to this post by Allan Heibein
Hi allan,
I want to know whether you successfully found the method to count stress granules/cell and quantification. I dont find any suitable method and strucking. It would be great if you share the procedure. Thanks |
«
Return to ImageJ
|
1 view|%1 views
| Free forum by Nabble | Edit this page |