*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy*****
Hi Claire,
I was fortunate enough to be able to attend a STORM workshop by the Xiang lab at Harvard university this past summer, and they did a very good job of taking the workshop attendants through the entire imaging process (sample prep, acquisition, image processing, etc.). As I recall from the processing session, the choice in the image pixel dimensions in the super-resolved image was determined in part by the precision of the Gaussian fitting that is used to pinpoint the location of a single molecule on their microscope system (and of course their are many factors that affect this, but mainly the number of photons registered per localization event).
Yes, the final super-resolved image is much bigger in file size because of these increased pixel dimensions and yes you most definitely keep the raw data. As an example, a typical raw STORM stack (one colour channel) consisted of anywhere from 10000 to 25000 images (512x512 pixels, 167 nm x 167 nm per pixel) and was approximately 2.6-3.0 GB in size. The rendered super-resolution images were somewhere on the order of 2138x2138 pixels and 17 MB in size. Of course, you always have the option to crop down during acquisition to a much smaller pixel region and this will reduce the file sizes; that is to say, you don't have to collect the full 512x512 frame if your are of interest is small compared to the larger field of view. Storage did not seem to be an issue there. Computer memory is cheap these days, so they likely have a devoted computer space just for this sort of thing. I didn't get any sense from them that they were concerned about the large GBs of data that were created on a daily basis. Perhaps emailing them directly would be the best thing to find out what specific image storage solutions they've decided on.
John Oreopoulos
On 2010-11-18, at 10:34 AM, Claire Brown wrote:
> *****
> To join, leave or search the confocal microscopy listserv, go to:
>
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy> *****
>
> I was wondering when people create PALM and STORM images how do they choose
> the pixel size in the high resolution images? Do people keep the "raw" data
> as well? This must create some major image storage and back up concerns.
>
> Sincerely,
>
> Claire
Locked
