Ion intensity for ROI

[Ramin Vismeh]

Hello all,
Does anyone know how we can get the average ion intensity for a ROI? I can see an spectra for it in Analysis—->Plot—->ROI, is that the average intensity of ions? what happens if the ion intensity is more than 32767?

Thanks

[David Pirman]

You should be able to click on ROI–>statistics. You must ensure that the N(m/z) values match the particular m/z image you are interested in. Actually I think there’s an error, where the N value in the stats window should be +1 from the image.

Dave Pirman

[Ramin Vismeh]

Thanks Dave for replying…
Yes, that makes sense…Do you know how we can get the area for the ROI?

Thanks,
Ramin

[David Pirman]

When you click on statistics, you should see a window with some bar graphs and tabulated data. One of the columns is says volume (mm^2). I think this is the area since its in squared units. There is also a column of population which should be the # averaged spectra. I assume if the two ROI’s have the same population#, they should be same area.
If this statement is wrong, hopefully someone can comment further.

Dave

[Ramin Vismeh]

Hi Dave,

Thats right. that is actually the number of pixels of ROI. But the thing is apparently it assumes each pixel is 1 mm^2. I guess the real size PROBABLY depends on the pixel size that you set up in your imaging experiment. Please correct me if I am wrong.

Ramin

[David Pirman]

Ramin,

I do remember something funny about the area when I was doing some quantitative work. I would typically measure my area by using my step size and laser diameter with a known amount of scans across the ROI, followed by a simple area calculation. This is what I would recommend.

There seems to be a few quirks with software when using it for quantitative applications. If I come across any more, I’ll post them on this forum.

Dave

[Ramin Vismeh]

[Ramin Vismeh]

Dave,

I am using DESI. So, my spot size is not known actually. I can have an estimation but it wont be accurate. Thats why I am trying to assess the area using the image. How do you usually generate the standard curves? I assume you use something like ion intensity vs. ng/area for some spots of standard that you put on tissues. Is that right? When you image the tissue, well most of the times the compound you are looking for is not probably in one nice circle spot. It is usually scattered all over the tissue. So, I assume for quantification the right unit to use is ng/area. Am I right?

Thanks

[David Pirman]

Ramin,

So for quantitation, it can be a bit tricky, but you are correct. You should definitely think about ng/area and then even volume (since you know your tissue thickness). I generate standard curves by spotting calibration standards beneath my tissue (in a fixed circular area) average the intensity within the spot and then calculate the volume of tissue. Then I would quantify with same exact area size somewhere else on the tissue, but I haven’t had much luck in this without an internal standard. So in my case, my calibration standards were in a nice confined area with a uniformly applied internal standard. If you are trying to generate quantitative data from a compounds with unknown area density information, I’d think it would be quite difficult since the calibration curve is heavily dependent on the calculation of area on the x-axis. I just had a paper accepted in Analytical Chemistry using this technique for quantitation.
For calculating the area accurately in DESI, I would suggest drawing an ROI and overlaying the image onto an optical image of your tissue and basically doing a manual measurement of the area you are interested in. Also, placing some distance registration marks on the MALDI target. I typically use glass slides, and will simply draw marks of interest on the backside so when I take an optical image I can make measurements and co-register the MS image.

Dave

[Author]

Posts