Re: colocalisation without software
Locked
 
|
Glen MacDonald-2 |
Re: colocalisation without software
|
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal Colocalization based upon "yellow" could be accurate, if and only if, the intensities are comparable and pixel (voxel) quantities in the suspected colocalized volumes are in roughly equal. . Otherwise, the yellow is masked by the predominate channel. Something small, like lysosomes, would need to be sampled properly. Colocalization could be masked by blur unless deconvolved, even if images are collected with a confocal. On Feb 7, 2007, at 1:05 PM, Marc Thibault wrote: > Search the CONFOCAL archive at > http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal > > Hi all, > > It seems that in many papers from biologists or chemists, and i'm > talking > high impact factors journals, colocalisation of two elements is is > often > assumed by simple color superposition (ex: red and green fluoresce > yellow > when colocalising), while microscopists (many physisists I suppose) > seem to > need a more complex software-based confirmation. > Is it ok, when using high end equipment and corrected objectives > (apochromat > with high NA for ex.), to assume colocalisation by color > superposition, > especially when fluorophore are confined to small volume entities, > like > lysosomes ? > > Thanks > > Marc |
|
|
M. van de corput |
Re: colocalisation without software
|
|
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal a good tutorial review by Bolte & Cordelieres J. Micr, 2006 is very helpful understanding co-localization analysis. Mariette ____________________________________________ Dr. M.P.C. Kemner-van de Corput, ____________________________________________ MGC - Dept. of Cell Biology & Genetics Erasmus Medical Center Dr. Molewaterplein 50, 3015 GE Rotterdam POB 2040, 3000 CA Rotterdam, The Netherlands Office: H-Ee751; tel: +31 10 704.3949 Lab: H-Ee710; tel lab: +31 10 704.3315 tel secr: +31 10 704.3169 ____________________________________________ http://www2.eur.nl/fgg/ch1/cellbiology/ http://www.thesis.kemner.biz/ ____________________________________________ Op Vr, 28 maart, 2008 10:54 pm, schreef Glen MacDonald: > Search the CONFOCAL archive at > http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal > > Colocalization based upon "yellow" could be accurate, if and only if, > the intensities are comparable and pixel (voxel) quantities in the > suspected colocalized volumes are in roughly equal. . Otherwise, > the yellow is masked by the predominate channel. Something small, > like lysosomes, would need to be sampled properly. Colocalization > could be masked by blur unless deconvolved, even if images are > collected with a confocal. > On Feb 7, 2007, at 1:05 PM, Marc Thibault wrote: > >> Search the CONFOCAL archive at >> http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal >> >> Hi all, >> >> It seems that in many papers from biologists or chemists, and i'm >> talking >> high impact factors journals, colocalisation of two elements is is >> often >> assumed by simple color superposition (ex: red and green fluoresce >> yellow >> when colocalising), while microscopists (many physisists I suppose) >> seem to >> need a more complex software-based confirmation. >> Is it ok, when using high end equipment and corrected objectives >> (apochromat >> with high NA for ex.), to assume colocalisation by color >> superposition, >> especially when fluorophore are confined to small volume entities, >> like >> lysosomes ? >> >> Thanks >> >> Marc > |
|
|
Jeremy Adler |
Re: colocalisation without software
|
|
In reply to this post by Glen MacDonald-2
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal Problems with establishing colocalisation by overlaying images using different layers of an RGB image 1) A change to the detector gain and fiddling in Photoshop changes the apparent colocalisation. 2) The appearance of pixels, where the intensities are similar but are low, is highly dependent on the setup of the monitor and on limitations of printing. 3) Interpretation is therefore highly subjective and Journals should demand a higher level of evidence. 4) Software for quantitation is widely available. Most of which generates a very standard set of measurements. It should be recognized that the quantitative and non quantitative methods are dependent upon having images of good quality and that are precisely aligned Jeremy Adler Cell Biology The Wenner-Gren Inst. Arrhenius Laboratories E5 Stockholm University Stockholm 106 91 Sweden -----Original Message----- From: Confocal Microscopy List on behalf of Glen MacDonald Sent: Fri 3/28/2008 22:54 To: [hidden email] Subject: Re: colocalisation without software > Search the CONFOCAL archive at > http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal > > Hi all, > > It seems that in many papers from biologists or chemists, and i'm > talking > high impact factors journals, colocalisation of two elements is is > often > assumed by simple color superposition (ex: red and green fluoresce > yellow > when colocalising), while microscopists (many physisists I suppose) > seem to > need a more complex software-based confirmation. > Is it ok, when using high end equipment and corrected objectives > (apochromat > with high NA for ex.), to assume colocalisation by color > superposition, > especially when fluorophore are confined to small volume entities, > like > lysosomes ? > > Thanks > > Marc |
|
|
Valeria Berno |
Re: colocalisation without software
|
|
In reply to this post by Glen MacDonald-2
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal Hi, This question just fit in perfectly on what I am trying to find out about colocalization. When and why do I need do deconvolve pictures collected with a confocal in order to be sure about my colocalization (or not colocalization) results? To be specific: I am working on pre and post-synaptic proteins. Thanks Valeria > Search the CONFOCAL archive at > http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal > > Colocalization based upon "yellow" could be accurate, if and only if, > the intensities are comparable and pixel (voxel) quantities in the > suspected colocalized volumes are in roughly equal. . Otherwise, > the yellow is masked by the predominate channel. Something small, > like lysosomes, would need to be sampled properly. Colocalization > could be masked by blur unless deconvolved, even if images are > collected with a confocal. > On Feb 7, 2007, at 1:05 PM, Marc Thibault wrote: > >> Search the CONFOCAL archive at >> http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal >> >> Hi all, >> >> It seems that in many papers from biologists or chemists, and i'm >> talking >> high impact factors journals, colocalisation of two elements is is >> often >> assumed by simple color superposition (ex: red and green fluoresce >> yellow >> when colocalising), while microscopists (many physisists I suppose) >> seem to >> need a more complex software-based confirmation. >> Is it ok, when using high end equipment and corrected objectives >> (apochromat >> with high NA for ex.), to assume colocalisation by color >> superposition, >> especially when fluorophore are confined to small volume entities, >> like >> lysosomes ? >> >> Thanks >> >> Marc > |
|
|
Julio Vazquez |
Re: colocalisation without software
|
|
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
=
Hi Valeria, Here are a few thoughts: Deconvolution improves your images in different ways: it removes out of focus blur (more so on widefield images), it reduces noise (widefield and confocal), improves signal to noise (widefield and confocal), and "tightens" the PSF (or the boundaries of the image of your objects). All these factors will improve your colocalization studies. For instance, with better signal to noise, you get sharper, cleaner intensity peaks for your objects... these will be easier to threshold or otherwise analyze for colocalization. Less noise in your images will also facilitate your colocalization studies. In addition to those practical considerations, a deconvolved image gives a more accurate 3-Dd representation of your objects, therefore leading to more accurate colocalization results. One important issue to consider is the PSF: the image of a point object is some sort of blurred 3-D football (the American, elongated type). Deconvolution will shrink the boundaries of this football to some extent, and make them crisper (better defined). Even so, the image of a point object is a larger 3-D object (or PSF). The size of the PSF determines the degree of uncertainty in the actual size and boundaries of your object. The error is larger in the vertical axis, and lesser in the horizontal directions. For example, with a high NA 60x/1.4 oil lens, the dimensions of the PASF will be in the order of 0.2 microns in x/y and 0.5 microns in z. From this you can see that if you have two molecules located 0.1 micron apart (let's say along the x axis, their deconvolved images will overlap to some extent, even though the molecules do not actually touch. If the two molecules are 0.2 or even 0.3 microns apart along the z axis, their images will also similarly appear to overlap. However, by analyzing the images carefully (and better so in 3-D), and knowing the imaging properties of your objective (the PSF), you could probably determine that the images you are seeing are images of very small point objects, could determine the actual location of those objects (center of intensity of the images), and conclude there is no overlap. On non deconvolved images, this would be harder to do because there will be more blur, and more uncertainty in the actual size and location of the images (because of more noise in the images). Finally, you can see that since the image of an object is generally inflated by a well-defined quantity (namely, by the radius of the PSF). Therefore, the smaller and the closer the objects of interest are, the greater the potential for error: the difference between the size of a GFP molecule and its image is proportionally much greater than the difference between the size of a nucleus and the size of its image, because the dimensions of the PSF are relatively small compared to the size of the nucleus, but is quite large compared to the size of a molecule. Therefore, image quality, deconvolution, and knowing the properties of the microscope (especially the PSF of the particular objective used), will be much more important when you are looking at colocalization of small, faint objects, located close to each other, and will be less critical when looking at large objects... Similarly, when you are looking at very small objects near the resolution limit, it is even more critical to use optimal sampling (have a pixel size that's about 2.5 times smaller than the (width at 50% max intensity of the) PSF). Again, a factor of two in pixel size will be proportionally huge when trying to analyze very small punctate objects or filaments of dimensions comparable to the PSF, while it will be less critical if you are looking at whole cells within a tissue, because in that case, the difference in pixel size is tiny compared to the size of the cell. Understanding the concepts of resolution, PSF, and sampling are critical to producing accurate quantitative colocalization (or other) data from microscopy images, especially of very small objects near the limit of resolution. I really recommend taking some time to read about these concepts, e.g. at this site: (search for PSF, resolution, and such) Hope this helps... Julio. -- Julio Vazquez Fred Hutchinson Cancer Research Center Seattle, WA 98109-1024 == On Mar 31, 2008, at 8:05 AM, Valeria Berno wrote:
|
|
|
Mayandi Sivaguru |
Re: colocalisation without software
|
|
In reply to this post by Valeria Berno
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
Valeria, my understanding is that you will be better off with deconvolving all your optical microscope data sets (widefield, confocal and etc) in general. With reference to colocalization analysis, you first sample the data following sequential scans (never simultaneous for the coloc analysis) Nyquist sampling in 3D (I would personally suggest a bit over sampling won't hurt, if you do not experience significant photobleaching), and then a deconvolution is a must with a plane by plane analysis. Deconvolution will not change a "non-cocolalizing" data points in to "colocalizing" data points. But it can be otherwise, a colocaizing data points in raw data could become in fact not colocalizing anymore after deconvolution. But the parameters affecting your conlusion greatly is at much before you deconve the data i.e., the sample preparation, fixation, blocking, selection of antibodies, fluorophores, scan parameters and so on. Shiv At 10:05 AM 3/31/2008, you wrote: Search the CONFOCAL archive at Microscopy Facility Manager 8, Institute for Genomic Biology University of Illinois at Urbana-Champaign 1206 West Gregory Dr. Urbana, IL 61801 USA Office: 217.333.1214 Fax: 217.244.2496 [hidden email] http://core.igb.uiuc.edu |
|
|
Glen MacDonald-2 |
Re: colocalisation without software
|
|
In reply to this post by Valeria Berno
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal First, when I sent that email on colocalization without software, I was actually cleaning out unfinished drafts in my email, with the intent of deleting it. But, this is a topic that always initiates a discussion. As recommended the paper by Boldt and Cordeliere is very good, as is Costes et al, 2004 for the most recent papers, both have refs to prior work. the former paper nicely summarizes different types of describing colocalization. The confocal will not completely reject out of focus light due to residual aberrations in the optical pathway. The sample is often overlooked as an element of the optical pathway and it will often contribute to spherical aberration and background. Deconvolution of confocal stacks will further reduce noise and effects of spherical aberration and scattering. Of course, it is not a substitute for proper sample preparation, adequate sampling and a well maintained instrument. Noise in your images will reduce the apparent colocalization by inserting random pixels above threshold, while factors contributing to background will raise the appearance of colocalization since the signals are spreading throughout the image. As mentioned in Jeremy's email, a number of other factors need to be considered. Above all, and as well described in Costes etal, controls for your labeling are essential to avoid mistaking such things as autofluorescence, bleedthrough or non-specific labeling as intensities representing components for which you are analysing. You are pushing the limits of optical resolution, so you will need to push the limits of good methodology. Regards, Glen Glen MacDonald Core for Communication Research Virginia Merrill Bloedel Hearing Research Center Box 357923 University of Washington Seattle, WA 98195-7923 USA (206) 616-4156 [hidden email] ************************************************************************ ****** The box said "Requires WindowsXP or better", so I bought a Macintosh. ************************************************************************ ****** On Mar 31, 2008, at 8:05 AM, Valeria Berno wrote: > Search the CONFOCAL archive at > http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal > > Hi, > > This question just fit in perfectly on what I am trying to find out > about > colocalization. > > When and why do I need do deconvolve pictures collected with a > confocal in > order to be sure about my colocalization (or not colocalization) > results? > > To be specific: I am working on pre and post-synaptic proteins. > > Thanks > > Valeria > > > >> Search the CONFOCAL archive at >> http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal >> >> Colocalization based upon "yellow" could be accurate, if and only if, >> the intensities are comparable and pixel (voxel) quantities in the >> suspected colocalized volumes are in roughly equal. . Otherwise, >> the yellow is masked by the predominate channel. Something small, >> like lysosomes, would need to be sampled properly. Colocalization >> could be masked by blur unless deconvolved, even if images are >> collected with a confocal. >> On Feb 7, 2007, at 1:05 PM, Marc Thibault wrote: >> >>> Search the CONFOCAL archive at >>> http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal >>> >>> Hi all, >>> >>> It seems that in many papers from biologists or chemists, and i'm >>> talking >>> high impact factors journals, colocalisation of two elements is is >>> often >>> assumed by simple color superposition (ex: red and green fluoresce >>> yellow >>> when colocalising), while microscopists (many physisists I suppose) >>> seem to >>> need a more complex software-based confirmation. >>> Is it ok, when using high end equipment and corrected objectives >>> (apochromat >>> with high NA for ex.), to assume colocalisation by color >>> superposition, >>> especially when fluorophore are confined to small volume entities, >>> like >>> lysosomes ? >>> >>> Thanks >>> >>> Marc >> |
|
|
John Oreopoulos |
Re: colocalisation without software
|
|
In reply to this post by Mayandi Sivaguru
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
Does anyone know of any freely available software that can deconvolve image data? I am only aware of one ImageJ deconvolution plugin that does a reasonable job under certain circumstances. I'd be interested to know if anyone has created any others. John Oreopoulos, BSc, PhD Candidate University of Toronto Institute For Biomaterials and Biomedical Engineering Centre For Studies in Molecular Imaging Tel: W:416-946-5022 On 31-Mar-08, at 3:02 PM, Mayandi Sivaguru wrote: Search the CONFOCAL archive at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal |
|
|
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
Hi Jon,
If your university/department has a license, you might find Matlab's deconvblind, deconvwnr, deconvreg, and deconvlucy functions useful. They require basic Matlab skill and some knowledge about how to generate appropriate input PSFs, but I've used them successfully a few times. Best, Nate Nathan O'Connor Graduate Student Physiology and Biophysics Weill Medical College of Cornell University NY, NY 10021 On Mon, Mar 31, 2008 at 3:40 PM, John Oreopoulos <[hidden email]> wrote:
|
|
|
Farid Jalali |
Re: colocalisation without software
|
|
In reply to this post by Valeria Berno
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
Journal of Microscopy, Vol. 208, Pt 2 November 2002, pp. 134–147
Deconvolution improves colocalization analysis of multiple fluorochromes in 3D confocal data sets more than filtering techniques L. LANDMANN Microscopy and Research Technique 64:103–112 (2004) Colocalization Analysis Yields Superior Results After Image Restoration LUKAS LANDMANN* AND PERMSIN MARBET Two nice papers that specifically address the issue at hand. I am doing a great deal of co-localization analysis and the image intensity based correlative methods such as Pearson or Overlap greatly benefit from image restoration (I am using 3D Blind). The correlations generally end up being greater and using the method of Van Steensel (JaCOP plugin, pixel shift a re-calculation of Pearson) yields a steeper slope as one channel is shifted in registration relative to the other. Object based methods will as well benefit from this as noise is reduced, allowing for better image segmentation. The latter can yield centroids and for well sampled images, the distances between centroids can be meaningful. Centroids from the red and green channel with a distance of 0um between them is quite suggestive that the object defined by the two channels are close to each other. This can also be done quite easily in 3D as well with 3D object counter from Image J. As someone else had mentioned, this does operate at the limits of optical resolution, but I have found it very useful to develop an idea how close to targets are to each other and it can strengthen the image intensity based correlative methods. JaCOP and the J.Microscopy article by Bolte and Cordelieres are excellent. Suzanne Bolte helped me with the centroid approach using Image J. On Mon, Mar 31, 2008 at 11:05 AM, Valeria Berno <[hidden email]> wrote: Hi, -- Farid Jalali MSc Senior Research Technician/ Lab Manager Dr. Robert Bristow Lab Applied Molecular Oncology Princess Margaret Hospital Toronto, Canada 416-946-4501 X4351 (Princess Margaret Hospital) 416-581-7754 STTARR at MaRS Building 416-581-7791 STTARR Microscopy Suite |
|
|
Jeremy Adler |
Re: colocalisation without software
|
|
In reply to this post by Jeremy Adler
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
further thoughts on colocalisation in the context of deconvolution 1) A problem in measuring colocalisation is image quality - an important assumption underlying the Pearson correlation coefficient is that the data, in our case pixel intensities, is accurate. As a simple test acquire an image and immediately collect a second image, compare the same pixels in the 2 images - they are very unlikely to be identical, due to Poisson and background noise. Noise creates a difference between the measured Pearson correlation and the 'noise free' Pearson correlation. However we can obtain a 'noise free' correlation by measuring and then factoring out image quality from the calculation.
Adler, Pagakis, Parmryd 2008
"Replicate based noise corrected correlation for accurate measurements of colocalization"
J. Microscopy Vol. 230, issue 1 (in press shortly) 2) The Bolte and Cordelieres review has been cited (J Microscopy, 206, 213-232) favourably and indeed it is a useful review, but it also shows a fundamental misunderstanding of both data acquisition and of correlation measurement. This is apparent in the scattergrams in their Fig 5, which clearly show that some of the images have pixels that are either saturated or not even onscale, but were still included in the calculation of the correlation (correspondence with the authors). This may explain the strange correlation coefficients: Fig 5c is, by eye, uncorrelated, the data points are all over the place, but it has a reported correlation of 0.69, which is only slightly lower than the 0.8? associated with Fig5b, which clearly has an appreciable correlation. Equally strange is Fig 5c with a negative correlation of -0.3, despite that fact that there seem to be almost no pixels in which both fluorophores occur - this isn't a negative correlation or even uncorrelation (a Pearson correlation coeff of 0) but simply the absence of any relationship - the two fluorophores are never in the same place. The major problem in their measurements of correlation is that no attempt was made to exclude pixels that are devoid of fluorescence (background only) and to limit the analysis to pixels actually containing both fluorophores. I must add that my explanation is slightly speculative, since the authors have refused permission for the renanalysis of their images, because we had criticised other aspects of the review (see ref). I would encourage everyone to examine Fig 5 and draw your own conclusions. Adler & Parmryd, J Microscopy, 2007, 227, page 83 a reply by Bolte and Cordelieres appears in the same issue Jeremy Adler Cell Biology The Wenner-Gren Inst. Arrhenius Laboratories E5 Stockholm University Stockholm 106 91 Sweden -----Original Message----- From: Confocal Microscopy List on behalf of Glen MacDonald Sent: Fri 3/28/2008 22:54 To: [hidden email] Subject: Re: colocalisation without software > Search the CONFOCAL archive at > http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal > > Hi all, > > It seems that in many papers from biologists or chemists, and i'm > talking > high impact factors journals, colocalisation of two elements is is > often > assumed by simple color superposition (ex: red and green fluoresce > yellow > when colocalising), while microscopists (many physisists I suppose) > seem to > need a more complex software-based confirmation. > Is it ok, when using high end equipment and corrected objectives > (apochromat > with high NA for ex.), to assume colocalisation by color > superposition, > especially when fluorophore are confined to small volume entities, > like > lysosomes ? > > Thanks > > Marc |
|
|
Jeremy Adler |
Re: colocalisation without software typo removed
|
|
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
my apologies for apparently resending this email, I noticed a significant typo further thoughts on colocalisation in the context of deconvolution 1) A problem in measuring colocalisation is image quality - an important assumption underlying the Pearson correlation coefficient is that the data, in our case pixel intensities, is accurate. As a simple test acquire an image and immediately collect a second image, compare the same pixels in the 2 images - they are very unlikely to be identical, due to Poisson and background noise. Noise creates a difference between the measured Pearson correlation and the 'noise free' Pearson correlation. However we can obtain a 'noise free' correlation by measuring and then factoring out image quality from the calculation.
Adler, Pagakis, Parmryd 2008
"Replicate based noise corrected correlation for accurate measurements of colocalization"
J. Microscopy Vol. 230, issue 1 (in press shortly) 2) The Bolte and Cordelieres review has been cited (J Microscopy, 206, 213-232) favourably and indeed it is a useful review, but it also shows a fundamental misunderstanding of both data acquisition and of correlation measurement. This is apparent in the scattergrams in their Fig 5, which clearly show that some of the images have pixels that are either saturated or not even onscale, but were still included in the calculation of the correlation (correspondence with the authors). This may explain the strange correlation coefficients: Fig 5c is, by eye, uncorrelated, the data points are all over the place, but it has a reported correlation of 0.69, which is only slightly lower than the 0.8? associated with Fig5b, which clearly has an appreciable correlation. Equally strange is Fig 5d with a negative correlation of -0.3, despite that fact that there seem to be almost no pixels in which both fluorophores occur - this isn't a negative correlation or even uncorrelation (a Pearson correlation coeff of 0) but simply the absence of any relationship - the two fluorophores are never in the same place. The major problem in their measurements of correlation is that no attempt was made to exclude pixels that are devoid of fluorescence (background only) and to limit the analysis to pixels actually containing both fluorophores. I must add that my explanation is slightly speculative, since the authors have refused permission for the renanalysis of their images, because we had criticised other aspects of the review (see ref). I would encourage everyone to examine Fig 5 and draw your own conclusions. Adler & Parmryd, J Microscopy, 2007, 227, page 83 a reply by Bolte and Cordelieres appears in the same issue Jeremy Adler Cell Biology The Wenner-Gren Inst. Arrhenius Laboratories E5 Stockholm University Stockholm 106 91 Sweden -----Original Message----- From: Confocal Microscopy List on behalf of Glen MacDonald Sent: Fri 3/28/2008 22:54 To: [hidden email] Subject: Re: colocalisation without software > Search the CONFOCAL archive at > http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal > > Hi all, > > It seems that in many papers from biologists or chemists, and i'm > talking > high impact factors journals, colocalisation of two elements is is > often > assumed by simple color superposition (ex: red and green fluoresce > yellow > when colocalising), while microscopists (many physisists I suppose) > seem to > need a more complex software-based confirmation. > Is it ok, when using high end equipment and corrected objectives > (apochromat > with high NA for ex.), to assume colocalisation by color > superposition, > especially when fluorophore are confined to small volume entities, > like > lysosomes ? > > Thanks > > Marc |
|
|
M. van de corput |
Re: colocalisation without software
|
|
In reply to this post by Jeremy Adler
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal Jeremy Adler wrote: > Search the CONFOCAL archive at > http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal > > further thoughts on colocalisation in the context of deconvolution > > 1) A problem in measuring colocalisation is image quality - an > important assumption underlying the Pearson correlation coefficient is > that the data, in our case pixel intensities, is accurate. As a > simple test acquire an image and immediately collect a second image, > compare the same pixels in the 2 images - they are very unlikely to be > identical, due to Poisson and background noise. importance of accurate image acquisition. I agree the image acquisition is the first and very important step towards any analysis done on an image. I use a 4 line scan and average the intensity of the 4 resulting in an average intensity image.....isn't that a way to reduce noise and determine the "left over noise/background" more accurately? It surely improves the image. Deconvolution then takes away a lot more noise as well. by using measured SNR for each channel, and using a measure PSF for each channel separately. After chromatic shift correction the image is ready for colocalization analysis. There will always be some noise left in the image as not all the noise is from equal intensity (due to e.g. varying autofluorescence of biological structures). I wonder if there is a list of do and do nots for doing colocalization studies (besides the helpful list in cited articles). Or better: what is allowed and what not on image processing. Mariette van de Corput Erasmus MC, Rotterdam, NL > > > > > -----Original Message----- > From: Confocal Microscopy List on behalf of Glen MacDonald > Sent: Fri 3/28/2008 22:54 > To: [hidden email] > Subject: Re: colocalisation without software > > > Search the CONFOCAL archive at > > http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal > > > > Hi all, > > > > It seems that in many papers from biologists or chemists, and i'm > > talking > > high impact factors journals, colocalisation of two elements is is > > often > > assumed by simple color superposition (ex: red and green fluoresce > > yellow > > when colocalising), while microscopists (many physisists I suppose) > > seem to > > need a more complex software-based confirmation. > > Is it ok, when using high end equipment and corrected objectives > > (apochromat > > with high NA for ex.), to assume colocalisation by color > > superposition, > > especially when fluorophore are confined to small volume entities, > > like > > lysosomes ? > > > > Thanks > > > > Marc |
«
Return to Confocal Microscopy List
|
1 view|%1 views
| Free forum by Nabble | Edit this page |