Thank Dr. Daly for the helpful tutorial. I need to do colocalisation analysis today and this is a very useful video. Please continue with the tutorial.
@@CraigDaly Thank you Dr. Daly! That would really help all students that are just starting to do research including myself. Looking forward to your new videos.
Great explanation, thank you. The reason the Cytofluorogram is not visible is that for some reason the color of the content is set to "white". On the plot window, go to "More>>", then "Contents style..." and set it to any other color. It is not your PC. Cheers.
Hi! Thank you for your video! I'm a PhD student at the UoG and I'm having some troubles with the colocalization study of my proteins. I've used Jacop as suggested in the video, but I'm having some toubles: 1) The tuneable thresholds it shows are not the ones computed with the costes algorithm and I don't understand how it computes them; 2) the costes thresholds that it then shows in the results are very low ( if the threshold proposed is 100 for a channel, the costes one is 5) and 3) he computes the pearson's coefficient without any threshold or only with the costes one, which doesn't make much difference since the threshold is so low. Do you know if there's a way, with this or with another plugin, to set a threshold and compute the pearson's coefficient for that threshold? and do you know another automatic way to impose a threshold? since maybe changing it every time is both time consuming and not very objective. sorry for having so many questions, and thank you in advance :)
Hi Camilla, I would need to run it and have a look. I’m not on campus this week but I’ll be in next week if you want to drop by with some images. I won’t leave contact details here but I’m easy to find. Email me at UofG and we can sort something out. Craig.
Great video Craig. I was wondering if there is a way to quantify cells (count cells) with colocalization. For example, I am measuring GFP expression in neurons (so quantifying neurons positive for GFP as well as a neuronal marker)?
sorry for the slow response. I wonder if you could threshold and segment the different channels. Then do analyse particles to get a cell count and particle count on your marker. Then do a binary AND of the two segmented images. The result should be only those cells that also have the marker. A further analyse particles will give you the cell number. Disclaimer - all of that is without seeing your images. C.
Thank's a lot for the video! I was just wondering: is it statistically significant to use different colocalisation thresholds for each pair of images? I assume that laser intensity of some pictures may differ, and 'one man's meet is another man's poison', the selected threshold might pick out more than necessary in other pictures. Thank you very much!
Hi, it's a good question. Ideally you keep everything the same but if you need to change the threshold then just be clear in any methods section you right. Sometimes it might be necessary - always a problem to wrestle with. Good luck.
Hi Craig, great video thanks. The JACoP plugin looks perfect for what I'm looking to do but crashes when I click 'analyse'. Should I be able to run this on a standard laptop or does it need something more powerful?
Hi, it should run fine on a laptop. Maybe check if your images are 16 or 32bit. If so convert to 8bit and see if that works. Also try both ImageJ and Fiji it might be a Java thing that’s more stable in one of those.
Hi! Thank you for your explication, but i have a question: It's possible to quantify the colocalization of a specific ROI in the image? How can I do that?
Hi, I do not know a way to do it in the JaCOP plugin. However, if you are using Fiji or can find the 'Colocalisation Threshold' plugin. it will allow you to measure colocalisation in a circular ROI defined in image A or B of a pair. I am just finishing a new video and the one after that will be on Image Correction factors for colocalisation. I will now add a short piece on ROI measurement since I see that this could be of use. In Fiji you can find the plugin at (Analyze/colocalization/colocalization threshold). Not sure why it doesn't work for a square/rectangular ROI though.
Even though two images are opened in image J After opening the JACoP plugin its saying "at least two images should be opened " written in red. Unable to proceed further.
I had the same issue and it was solved once both images were converted to 8-bit images. My originals were 32-bit and the plugin didn't recognize them despite being open. Cheers.
Dear Craig, thank you very much for this explanatory video! My question is about the statistical analysis of the colocalization in Jacop. Which values should we use? Pearson's coefficient or Mander's coefficient? If you think Mander's, then which one is better: M1 or M2? Original or threshold values? Once again, thank you so much!
Hi, really difficult to say without knowing the experimental detail and requirement. It really depends. However, I would say Pearsons first and foremost with a threshold if background signal is high/problematic. If you need additional information then incorporate Manders. C.
Hi, I cant think of a specific paper that uses % area but I do recommend reading this general paper which covers far more than I do in this video; imaging.gurdon.cam.ac.uk/ext/files/Bolte_Cordelieres_2006.pdf
*Hello, Many Thanks for this helpful tutorial. Would you please share the used plugins for downloading (can be found in the plugins folder of the app)? Because it is impossible to find them on the official site (under maintenance and no one knows until when). I tried to find these plugins from other sources but impossible to finder the latest version, and when I find some oldest version, contents are corrupted so they were not working with me.*
Hi, I could try and email the plugin to you (if I can lift it off my work PC remotely). Go to my website at www.cardiovascular.org to get one of my contact email addresses and drop me an email. I’m obviously reluctant to leave a valid email address on a TH-cam channel. Regards. Craig.
Hi Craig, Great presentation and very useful. I am not able to use these plugins for selected ROIs. Could you please suggest me something about colocalization of selected ROI? Because I think a lot of empty black area of an image and less area under the cells would deviate the colocalization R values. Thank you.
Hi, I’m not aware of any plugins that will do colocalisation on ROIs specifically. I have always just used JaCoP. All I can suggest is working through the multitude of colocalisation plugins available. Maybe one of the subscribers could pitch in. Anyone? Having said that, does JaCoP not work if you use the Costes Threshold option to mask out the background? Otherwise, maybe just crop/edit the images first? Sorry, not the answer you were hoping for I’m sure. C
Hi, I have found that JACoP BIOP version does analysis on ROIs! It is essentially JACoP with this added functionality, you just have to use a multi-channel image for it and then define which channels to use. You can then select between Costes automatic threshold, one of the ImageJ automatic ones or manually set your own. It is part of the BIOP library!
Hello, I am trying to analyze the concentration of protein in the nuclei and I'm confused on how to use the thereshold. If I have different splits to analise would I need to select a thereshold for each of them? Thank you beforehand
Hi, sorry for the delay. Yes, thresholding at different levels might help. Try applying a 6 colour LUT to ‘bin’ the image into 6 intensity ranges. That might help you figure out the next step. C.
Hello, thank you very much for this useful tutorial. Could you please share the plugins used to download, I can't find it anywhere. Thank you very much in advance
Hi. Two possibilities. You can email me at University of Glasgow and I will send the plugin. Alternatively, use Fiji and load the 'BIOP' suite of plugins. That includes a very nice version of JaCOP. You can find an excellent video on that plugin (and installation) here; th-cam.com/video/dk3ETh8oSX0/w-d-xo.html (by Johanna M. Dela Cruz) Good luck. Craig.
Hi Anny, try this; www.google.co.uk/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&ved=2ahUKEwi15OiGiLDoAhVBhlwKHT8qBIAQFjABegQIBhAB&url=https%3A%2F%2Fimagej.nih.gov%2Fij%2Fplugins%2Ftrack%2Fjacop2.html&usg=AOvVaw3x2u-C4dyrzjzTxSuTtEl8
Hi, thank you for this really useful tutorial. Could you please share the plugins used to download for JACoP, the link shared in the comments says its corrupted, any help would be great!
Hi Imogen, there is a better way now. Use Fiji and get the 'BIOP' plug in. Once installed, go to the analysis section and you will see BIOP JaCOP. This seems to be a slightly more stable and user friendly approach.
I would say that if all images were collected under identical conditions then yes you should use the same threshold value for all. However, if fluorescence varies a lot and conditions were slightly different (maybe detector sensitivities) then you may need to use an automatic threshold and then just explain what you did and why. Short answer; it depends on the experiment.
I have an issue. When I try to adjust the image threshold i don't get the bars in yellow color, it only has a little green bar so I cannot select the possible colocalisation dots. It depends on the format image or something?
I think it depends on the pixels count at the color region, to some extent the abundance of each color in the whole image, I'm not sure, but I observed similar things like you.
For some reason my JACoP is not reading my images. I opened them befor JACoP and it still says "No Image" on both fields. Reinstalled already JACoP, ImageJ and Fiji and nothing fixed. Anyone have this issue?
Hi, sorry for the slow response, TH-cam seems to have stopped alerting me. Try using Fiji and the plugin BIOP, this has an installation of JaCOP which is a bit more stable. C
Thank you for your great videos. I have two questions regarding the technique. Is it necessary to have the whole cell in the picture? Or is it more important to use a small pixel size? We want to measure the colocalization of a protein with mitochondria. And is deconvolution necessary? Best regards Karoline
Hi Karoline, No, not necessary to have the whole cell if you are just analysing the mitochondria. Deconvolution? Good question. If both channels can be deconvolved I think that would give a more accurate colocalization measurement. Difficult to say without seeing the original images. I would recommend the noise correction RBNCC method in my other video. Thanks for your comment. Craig.
@@CraigDaly Hi. Could you please let me know what is the name of the method you used in the first part of the analysis using Area method (percentage colocalization).
@@CraigDaly Thanks for the response. I am using just the basic percent colocalization and do not plan to do correlations tests. As I asked you earlier, is it possible for you to at least give me a reference that I can cite that uses the area (percent colocalization) method?
Thank Dr. Daly for the helpful tutorial. I need to do colocalisation analysis today and this is a very useful video. Please continue with the tutorial.
Thanks Cristina. I’m thinking of doing a part 2 for the colocalisation tutorial that will incorporate correcting for image noise prior to measurement.
@@CraigDaly Thank you Dr. Daly! That would really help all students that are just starting to do research including myself. Looking forward to your new videos.
Fantastic Video Craig! Thanks
Thanks for taking the time to comment Jonas 👍
Thank you so much for your video.
very helpful, thank you!
Very helpful! Thank you! :)
Really helpful video, thanks Craig! Any references that you've followed and can be cited for Colocalization?
Thank you so much!
Thank you !!
thanks Craig
Great explanation, thank you. The reason the Cytofluorogram is not visible is that for some reason the color of the content is set to "white". On the plot window, go to "More>>", then "Contents style..." and set it to any other color. It is not your PC. Cheers.
Ahh, thanks for that Atticus. Great tip. 👍
Hi! Thank you for your video! I'm a PhD student at the UoG and I'm having some troubles with the colocalization study of my proteins. I've used Jacop as suggested in the video, but I'm having some toubles: 1) The tuneable thresholds it shows are not the ones computed with the costes algorithm and I don't understand how it computes them; 2) the costes thresholds that it then shows in the results are very low ( if the threshold proposed is 100 for a channel, the costes one is 5) and 3) he computes the pearson's coefficient without any threshold or only with the costes one, which doesn't make much difference since the threshold is so low.
Do you know if there's a way, with this or with another plugin, to set a threshold and compute the pearson's coefficient for that threshold? and do you know another automatic way to impose a threshold? since maybe changing it every time is both time consuming and not very objective.
sorry for having so many questions, and thank you in advance :)
Hi Camilla, I would need to run it and have a look. I’m not on campus this week but I’ll be in next week if you want to drop by with some images. I won’t leave contact details here but I’m easy to find. Email me at UofG and we can sort something out. Craig.
Great video Craig. I was wondering if there is a way to quantify cells (count cells) with colocalization. For example, I am measuring GFP expression in neurons (so quantifying neurons positive for GFP as well as a neuronal marker)?
sorry for the slow response. I wonder if you could threshold and segment the different channels. Then do analyse particles to get a cell count and particle count on your marker. Then do a binary AND of the two segmented images. The result should be only those cells that also have the marker. A further analyse particles will give you the cell number. Disclaimer - all of that is without seeing your images. C.
Thank's a lot for the video! I was just wondering: is it statistically significant to use different colocalisation thresholds for each pair of images? I assume that laser intensity of some pictures may differ, and 'one man's meet is another man's poison', the selected threshold might pick out more than necessary in other pictures. Thank you very much!
Hi, it's a good question. Ideally you keep everything the same but if you need to change the threshold then just be clear in any methods section you right. Sometimes it might be necessary - always a problem to wrestle with. Good luck.
Hi Craig, great video thanks. The JACoP plugin looks perfect for what I'm looking to do but crashes when I click 'analyse'. Should I be able to run this on a standard laptop or does it need something more powerful?
Hi, it should run fine on a laptop. Maybe check if your images are 16 or 32bit. If so convert to 8bit and see if that works. Also try both ImageJ and Fiji it might be a Java thing that’s more stable in one of those.
Thank you, I will give it a try!
Hi! Thank you for your explication, but i have a question: It's possible to quantify the colocalization of a specific ROI in the image? How can I do that?
Hi, I do not know a way to do it in the JaCOP plugin. However, if you are using Fiji or can find the 'Colocalisation Threshold' plugin. it will allow you to measure colocalisation in a circular ROI defined in image A or B of a pair. I am just finishing a new video and the one after that will be on Image Correction factors for colocalisation. I will now add a short piece on ROI measurement since I see that this could be of use. In Fiji you can find the plugin at (Analyze/colocalization/colocalization threshold). Not sure why it doesn't work for a square/rectangular ROI though.
@@CraigDaly thank you very much Dr.!!!
Even though two images are opened in image J After opening the JACoP plugin its saying "at least two images should be opened " written in red. Unable to proceed further.
I had the same issue and it was solved once both images were converted to 8-bit images. My originals were 32-bit and the plugin didn't recognize them despite being open. Cheers.
Dear Craig, thank you very much for this explanatory video! My question is about the statistical analysis of the colocalization in Jacop. Which values should we use? Pearson's coefficient or Mander's coefficient? If you think Mander's, then which one is better: M1 or M2? Original or threshold values? Once again, thank you so much!
Hi, really difficult to say without knowing the experimental detail and requirement. It really depends. However, I would say Pearsons first and foremost with a threshold if background signal is high/problematic. If you need additional information then incorporate Manders. C.
Thank you for the video! Could you please provide a publication that talks about/uses the percent area of colocalization? Thank you!
Hi, I cant think of a specific paper that uses % area but I do recommend reading this general paper which covers far more than I do in this video;
imaging.gurdon.cam.ac.uk/ext/files/Bolte_Cordelieres_2006.pdf
*Hello, Many Thanks for this helpful tutorial. Would you please share the used plugins for downloading (can be found in the plugins folder of the app)? Because it is impossible to find them on the official site (under maintenance and no one knows until when). I tried to find these plugins from other sources but impossible to finder the latest version, and when I find some oldest version, contents are corrupted so they were not working with me.*
Hi, I could try and email the plugin to you (if I can lift it off my work PC remotely). Go to my website at www.cardiovascular.org to get one of my contact email addresses and drop me an email. I’m obviously reluctant to leave a valid email address on a TH-cam channel. Regards. Craig.
@@CraigDaly Thank you, just sending you an email on c***.org, mine si starting with B**. Kind regards
Hi Craig, Great presentation and very useful. I am not able to use these plugins for selected ROIs. Could you please suggest me something about colocalization of selected ROI? Because I think a lot of empty black area of an image and less area under the cells would deviate the colocalization R values. Thank you.
Hi, I’m not aware of any plugins that will do colocalisation on ROIs specifically. I have always just used JaCoP. All I can suggest is working through the multitude of colocalisation plugins available. Maybe one of the subscribers could pitch in. Anyone? Having said that, does JaCoP not work if you use the Costes Threshold option to mask out the background? Otherwise, maybe just crop/edit the images first? Sorry, not the answer you were hoping for I’m sure. C
Hi, I have found that JACoP BIOP version does analysis on ROIs! It is essentially JACoP with this added functionality, you just have to use a multi-channel image for it and then define which channels to use. You can then select between Costes automatic threshold, one of the ImageJ automatic ones or manually set your own. It is part of the BIOP library!
@@PurpleFlowerAngela Thank you so much for the infrmation. It sounds great and very useful. I will go through it.
@@PurpleFlowerAngela Thanks. I only just discovered the BIOP version a few weeks ago. It's a big improvement on the original plug in.
@@PurpleFlowerAngela Thank you very much for sharing this information, it was very useful.
Hello, I am trying to analyze the concentration of protein in the nuclei and I'm confused on how to use the thereshold. If I have different splits to analise would I need to select a thereshold for each of them? Thank you beforehand
Hi, sorry for the delay. Yes, thresholding at different levels might help. Try applying a 6 colour LUT to ‘bin’ the image into 6 intensity ranges. That might help you figure out the next step. C.
Hello, thank you very much for this useful tutorial. Could you please share the plugins used to download, I can't find it anywhere.
Thank you very much in advance
Hi. Two possibilities. You can email me at University of Glasgow and I will send the plugin. Alternatively, use Fiji and load the 'BIOP' suite of plugins. That includes a very nice version of JaCOP. You can find an excellent video on that plugin (and installation) here;
th-cam.com/video/dk3ETh8oSX0/w-d-xo.html (by Johanna M. Dela Cruz)
Good luck. Craig.
JACoP isn't available to download in imageJ. Is there another source where I can get this plugin?
Hi Anny, try this; www.google.co.uk/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&ved=2ahUKEwi15OiGiLDoAhVBhlwKHT8qBIAQFjABegQIBhAB&url=https%3A%2F%2Fimagej.nih.gov%2Fij%2Fplugins%2Ftrack%2Fjacop2.html&usg=AOvVaw3x2u-C4dyrzjzTxSuTtEl8
is that possible to quantifie immunofluoresence from images directly with this Apple?
Hi, yes. The process is measuring brightness in the image. The source of the brightness would not matter. Immunofluorescent images should work fine.
@@CraigDaly thank you
Hi, thank you for this really useful tutorial. Could you please share the plugins used to download for JACoP, the link shared in the comments says its corrupted, any help would be great!
Hi Imogen, there is a better way now. Use Fiji and get the 'BIOP' plug in. Once installed, go to the analysis section and you will see BIOP JaCOP. This seems to be a slightly more stable and user friendly approach.
05:00 I still do not realize how correctly adjust threshold, if I ve many of images. Do Ive to use always same permanent threshold?
I would say that if all images were collected under identical conditions then yes you should use the same threshold value for all. However, if fluorescence varies a lot and conditions were slightly different (maybe detector sensitivities) then you may need to use an automatic threshold and then just explain what you did and why. Short answer; it depends on the experiment.
@@CraigDaly Thank you very much for prompt reply and useful info!
I have an issue. When I try to adjust the image threshold i don't get the bars in yellow color, it only has a little green bar so I cannot select the possible colocalisation dots. It depends on the format image or something?
Ok, even when the software doesn't show me the bars, I selected only the yellow dots. Is this way also right?
Hi, have you converted the image to type RGB?
I think it depends on the pixels count at the color region, to some extent the abundance of each color in the whole image, I'm not sure, but I observed similar things like you.
@@eagleeagle4417 I think that's the reason. It doesn't matter if you select only the yellow color so you can continue with the analysis. Thank you.
Hi! Thank you for the video. Can you share the link for the Jacop plugin for iOS please
Hi, there is a better way to use JaCoP now. Use Fiji and find the BIOP plugin. It has an installation of JaCoP that seems to be a bit more stable. C
Thank you! Anybody encountering the same problem as me where there is no JACoP plugin at all?!
Hi, try downloading from here; imagejdocu.tudor.lu/doku.php?id=plugin:analysis:jacop_2.0:just_another_colocalization_plugin:start
@@CraigDaly Thank you so much!
For some reason my JACoP is not reading my images. I opened them befor JACoP and it still says "No Image" on both fields. Reinstalled already JACoP, ImageJ and Fiji and nothing fixed. Anyone have this issue?
Hi, sorry for the slow response, TH-cam seems to have stopped alerting me. Try using Fiji and the plugin BIOP, this has an installation of JaCOP which is a bit more stable. C
@@CraigDaly in the End was the format. It was not in RGB. When I put in the format It worked for me. Thanks
Sir how to convert the 8bit image to a colored image
Image/Type/RGB colour
Thank you for your great videos. I have two questions regarding the technique. Is it necessary to have the whole cell in the picture? Or is it more important to use a small pixel size? We want to measure the colocalization of a protein with mitochondria. And is deconvolution necessary?
Best regards
Karoline
Hi Karoline, No, not necessary to have the whole cell if you are just analysing the mitochondria. Deconvolution? Good question. If both channels can be deconvolved I think that would give a more accurate colocalization measurement. Difficult to say without seeing the original images. I would recommend the noise correction RBNCC method in my other video. Thanks for your comment. Craig.
@@CraigDaly Thank you for your answer!
@@CraigDaly Hi. Could you please let me know what is the name of the method you used in the first part of the analysis using Area method (percentage colocalization).
@@pratikkatwal7724 hi, it has no specific name that I know of. C
@@CraigDaly Thanks for the response. I am using just the basic percent colocalization and do not plan to do correlations tests. As I asked you earlier, is it possible for you to at least give me a reference that I can cite that uses the area (percent colocalization) method?