FISH measurement

[Boswell, Carl A - (cboswell)]

Hi all,
I have a user who wants to measure the volume, not intensity, of the
FISH-labeled region in a z-series.

Any issues with the following proposal, besides using a histogram to get
pixel numbers?

1. Threshold the FISH signals for control and experimental Z-Series
images with identical conditions.
2. Create a histogram for each thresholded image and use total pixel number
as
the FISH signal volume.

Thanks,
C

Carl A. Boswell, Ph.D.
Molecular and Cellular Biology
University of Arizona
520-954-7053
FAX [phone]

[JOEL B. SHEFFIELD]

It seems to me that you first have to be sure that each z-slice is independent --that you are not detecting fluorescence from one slice in another.--
Joel

On Thu, Aug 20, 2009 at 2:31 PM, Carl Boswell <[hidden email]> wrote:

Hi all,
I have a user who wants to measure the volume, not intensity, of the FISH-labeled region in a z-series.

Any issues with the following proposal, besides using a histogram to get pixel numbers?

1. Threshold the FISH signals for control and experimental Z-Series
images with identical conditions.
2. Create a histogram for each thresholded image and use total pixel number as
the FISH signal volume.

Thanks,
C

Carl A. Boswell, Ph.D.
Molecular and Cellular Biology
University of Arizona
520-954-7053
FAX [phone]

--


Joel B. Sheffield, Ph.D
Department of Biology
Temple University
Philadelphia, PA 19122
Voice: 215 204 8839
e-mail: [hidden email]
URL:  http://astro.temple.edu/~jbs

[Craig Brideau]

Sounds reasonable; in effect you are counting how many voxels are
above a particular threshold.  Once you have the number of voxels, if
you need an actual measurement just multiply by the real-world volume
of one voxel.

Craig


On Thu, Aug 20, 2009 at 12:31 PM, Carl
Boswell<[hidden email]> wrote:

> Hi all,
> I have a user who wants to measure the volume, not intensity, of the
> FISH-labeled region in a z-series.
>
> Any issues with the following proposal, besides using a histogram to get
> pixel numbers?
>
> 1. Threshold the FISH signals for control and experimental Z-Series
> images with identical conditions.
> 2. Create a histogram for each thresholded image and use total pixel number
> as
> the FISH signal volume.
>
> Thanks,
> C
>
> Carl A. Boswell, Ph.D.
> Molecular and Cellular Biology
> University of Arizona
> 520-954-7053
> FAX [phone]
>

[John Oreopoulos]

Correct me if I'm wrong, but in order to properly sample the image without aliasing (ie: proper Nyquist sampling), must you not set the lateral pixel and z-step size such that a single diffraction limited spot is sampled 2.3 times in xy and in z as well (bearing in mind that the resolution - counting just the voxels in that case would overestimate the volume, wouldn't it?

If you purposely set the voxel size to by 2 Airy units instead of 1, then you might avoid this problem, but you risk losing spatial resolution and creating aliasing artifacts, right? Perhaps this is okay if you don't care about the loss of spatial resolution?

John Oreopoulos

On 20-Aug-09, at 3:11 PM, JOEL B. SHEFFIELD wrote:

It seems to me that you first have to be sure that each z-slice is independent --that you are not detecting fluorescence from one slice in another.--
Joel

On Thu, Aug 20, 2009 at 2:31 PM, Carl Boswell <[hidden email]> wrote:

Hi all,
I have a user who wants to measure the volume, not intensity, of the FISH-labeled region in a z-series.

Any issues with the following proposal, besides using a histogram to get pixel numbers?

1. Threshold the FISH signals for control and experimental Z-Series
images with identical conditions.
2. Create a histogram for each thresholded image and use total pixel number as
the FISH signal volume.

Thanks,
C

Carl A. Boswell, Ph.D.
Molecular and Cellular Biology
University of Arizona
520-954-7053
FAX [phone]

--


Joel B. Sheffield, Ph.D
Department of Biology
Temple University
Philadelphia, PA 19122
Voice: 215 204 8839
e-mail: [hidden email]
URL:  http://astro.temple.edu/~jbs

[McDonald, David L]

Carl,

I believe the ImageJ 3D Objects Counter plugin will give the volume
of each object after thresholding the image.

Dave

Dave McDonald
Scientific Imaging Lab
Fred Hutchinson Cancer Research Center
1100 Fairview Avenue North, DE-512
Seattle, WA 98109
206-667-4205
http://www.fhcrc.org


At 11:31 AM 8/20/2009, you wrote:

>Hi all,
>I have a user who wants to measure the volume, not intensity, of the
>FISH-labeled region in a z-series.
>
>Any issues with the following proposal, besides using a histogram to
>get pixel numbers?
>
>1. Threshold the FISH signals for control and experimental Z-Series
>images with identical conditions.
>2. Create a histogram for each thresholded image and use total pixel number as
>the FISH signal volume.
>
>Thanks,
>C
>
>Carl A. Boswell, Ph.D.
>Molecular and Cellular Biology
>University of Arizona
>520-954-7053
>FAX [phone]

[John Oreopoulos]

Sorry, part of my message got cut off there, I was trying to say, ("bearing in mind that the resolution in z will be less than in xy").

John Oreopoulos

 

On 20-Aug-09, at 3:38 PM, John Oreopoulos wrote:

Correct me if I'm wrong, but in order to properly sample the image without aliasing (ie: proper Nyquist sampling), must you not set the lateral pixel and z-step size such that a single diffraction limited spot is sampled 2.3 times in xy and in z as well (bearing in mind that the resolution - counting just the voxels in that case would overestimate the volume, wouldn't it?

If you purposely set the voxel size to by 2 Airy units instead of 1, then you might avoid this problem, but you risk losing spatial resolution and creating aliasing artifacts, right? Perhaps this is okay if you don't care about the loss of spatial resolution?

John Oreopoulos

On 20-Aug-09, at 3:11 PM, JOEL B. SHEFFIELD wrote:

It seems to me that you first have to be sure that each z-slice is independent --that you are not detecting fluorescence from one slice in another.--
Joel

On Thu, Aug 20, 2009 at 2:31 PM, Carl Boswell <[hidden email]> wrote:

Hi all,
I have a user who wants to measure the volume, not intensity, of the FISH-labeled region in a z-series.

Any issues with the following proposal, besides using a histogram to get pixel numbers?

1. Threshold the FISH signals for control and experimental Z-Series
images with identical conditions.
2. Create a histogram for each thresholded image and use total pixel number as
the FISH signal volume.

Thanks,
C

Carl A. Boswell, Ph.D.
Molecular and Cellular Biology
University of Arizona
520-954-7053
FAX [phone]

--


Joel B. Sheffield, Ph.D
Department of Biology
Temple University
Philadelphia, PA 19122
Voice: 215 204 8839
e-mail: [hidden email]
URL:  http://astro.temple.edu/~jbs

[Mark Cannell]

Due to the non symmetrical PSF I don't think this is directly possible.
There is a solution tho' please see:

Soeller, C. & Cannell, M.B. (1999) Examination of the transverse tubular
system in living cardiac rat myocytes by 2-photon microscopy and digital
image-processing techniques. Circ. Res. 84: 266-275

Note that  this blurring method does not depend on imaging modality but
makes the effective PSF spherical which is what you need.

Cheers Mark


Carl Boswell wrote:

> Hi all,
> I have a user who wants to measure the volume, not intensity, of the
> FISH-labeled region in a z-series.
>
> Any issues with the following proposal, besides using a histogram to
> get pixel numbers?
>
> 1. Threshold the FISH signals for control and experimental Z-Series
> images with identical conditions.
> 2. Create a histogram for each thresholded image and use total pixel
> number as
> the FISH signal volume.
>
> Thanks,
> C
>
> Carl A. Boswell, Ph.D.
> Molecular and Cellular Biology
> University of Arizona
> 520-954-7053
> FAX [phone]