Co-localisation Analysis - immunofluorescence
Posted 15 November 2018 - 05:50 PM
Hoping someone can help. I need to do some colocalisation analysis for my IF images. I have Lamp2 staining in green and my protein X in red. I want to quantitate any changes in localisation between different treatments.
I would like to use Image J as I have access to this on my laptop.
Could someone give me a step by step guide on how to achieve this? Please don?t send me links to papers or plug-ins - I have not been able to use anything people have provided me so far. I can provide sample images if someone can help or can liaise with anyone who can help directly if necessary?
Thanks so much in advance
Posted 16 November 2018 - 07:07 AM
I would recommend getting Fiji - it is a variant of ImageJ that comes pre-packaged with a bunch of image analysis tools and works just the same. Base ImageJ is not much use for analysis unless you just want to count number of cells where you see the changes.
Having said that quantitation of images is really hard to do to get real numbers out - you need to ensure that the signal is strong enough that you can detect these changes somehow (computationally) and decide on what the changes look like (where is your cut-off for not changed?)
As you have given us no information on where you expect the colocalization to be in the cell and how big a change you are expecting to see, it is a little difficult to say exactly how to do an analysis. The sample images would be really helpful.
Posted 16 November 2018 - 07:18 PM
Hi Bob 1
I do have Fiji.
I would like to
1. count number of discrete protein x puncta (red)
2. measure colocalisation of protein x to lamp2 (green)
I have tried using the coloc2 feature but keep getting an error message of zero:zero ratio being too high. Also how do I count number of puncta?
There should be strong colocalisation of protein x to lamp2. In my images, I unexpectedly did see a change in this localization with less discrete puncta and less lamp2 colocalisation. I would therefore like to quantitate the changes I seem to see by eye.
Any step by step instructions would be really great. I tried to attach the czi file of the image but its telling me I'm not allowed to attach this type of file. Not sure why.
Posted 26 November 2018 - 09:08 AM
Ok, can you convert the czi image to a basic image file type such as png or jpg - the browsers can display this, but they can't display czi images, which is specific to Zeiss imaging systems.
Have a read of the following pages: https://imagej.net/C...zation_Analysis https://imagej.net/Coloc_2 and see if they are any help to start with. Colocalization is a really tricky thing to analyze properly, and I haven't done it in years, so I might not be much help.