Front Page › Forums › BioMap › Histogram function in BioMap
- This topic has 20 replies, 3 voices, and was last updated 13 years, 1 month ago by Ramin Vismeh.
-
AuthorPosts
-
October 28, 2011 at 7:26 pm #1378Ramin VismehParticipant
All,
Does anyone know how can I get a list for ion intensites of different ROIs in BioMap that can be exported to excel?
I have red this in one of Markus’s (Markus Stoeckli) papers but I cant figure what
exactly it says?“The data was loaded into BioMap, where regions-ofinterest
(ROIs) were drawn based on the optical scan to cover
the stomach, lung and trachea of the animal. These ROIs were
copied to the co-registered WBAL and MSI data sets. Using the
histogram function with 200 bins, the ROI data was reduced
and exported to Excel.”I can get the statistics of my ROIs but I am not sure how can I get a list that says what do I have in each pixel? That way I can remove empty pixles from the data set.
Thanks,
RaminNovember 4, 2011 at 12:21 pm #1386Markus StoeckliModerator[quote]Does anyone know how can I get a list for ion intensities of different ROIs in BioMap that can be exported to excel?[/quote]
For this, select “Roi” – “Statistics” and in there, select “File” – “Export” – “List”. The generated text file can directly be imported in Excel.[quote]I am not sure how can I get a list that says what do I have in each pixel?[/quote]
This is the same as exporting an images. You may do this in Analyze format or bmp.[quote]That way I can remove empty pixles from the data set.[/quote]
There is a function to remove “empty” pixels from ROIs in BioMap. Once you have drawn the ROIs, set the Min and Max intensity sliders to the intensity range you want to include. By selecting the grey color scale with blue on the low end and red on the top end one can easily select the matching values. Then select “Roi” – “Clip” and the pixels outside the selected intensity ranges will be removed from the ROIs. Redo the ROI statistics.November 7, 2011 at 6:12 pm #1392Ramin VismehParticipantThanks so much Markus for the Reply.:)
The other thing is the intensity of ions. I understand that the 32767 is the limit for parameters like data points. That’s why when creating an image; I use a narrow range of masses so that I end up with less # of data points. But I am not sure what BioMap does to the intensities. Is it going to cut the intensities that are above 32767 or is it dividing them all by a number so that the highest one becomes 32767 and scales the rest of them with the same ratio? I am talking about the numbers in Analysis—>plot—->ROI panel. I know when I look at my raw data, the intensities are higher than 32767 but when I generate ROIs, I get different intensities in the ROIs plot! My goal is to see if I really can trust the intensities in the ROI plot for ion intensities and use them as a basis for Quantification!Thanks,
RaminNovember 7, 2011 at 7:42 pm #1393Dodge BaluyaParticipantHi Ramin,
I believe the intensities are stored as single precision floats thus is not limited to 32767. (see the data structure, please correct me if I’m mistaken)
Think of the data points as containers for intensities values and the container are labelled with m/z. You can only work with a max of 32767 containers at one instance.You have to be careful when comparing your raw data with the Highlighted ROI data as most of the time ROI data is averaged according to the area you specified and the raw data is not.
Hope this helps.
Dodge
November 7, 2011 at 8:44 pm #1394Markus StoeckliModerator[b]dodge wrote:[/b]
[quote]…intensities are stored as single precision floating point…ROI data is averaged…[/quote]Thanks, I can confirm this. Markus
November 7, 2011 at 8:55 pm #1395Ramin VismehParticipantDodge and Markus,
Thanks. So you guys think that the intensities from ROI regions CAN be used as a base for quantification as they are not cut or scaled?
Dodge,
WhenI say the intensities do not match, I am talking about the average. When I average my raw data (say for the whole image) for a specific m/z, teh average is most of the times more than 32767! But if I select the whole image as one ROI, teh intensity for that specific m/z would not be the same in the ROI intensity panel! Is’nt the intensity divided by the number of pixels for the ROIs? another words: the ion intensity that I get from Analysys—>Plot—>ROI is the ion intensity/pixel?Thanks
November 7, 2011 at 9:42 pm #1396Dodge BaluyaParticipantLet me get more details to help solve this problem, I’m concerned that if this is true then this needs to be addressed as soon as possible.
How do you average your raw data for the whole image (non-ROI method)?Analysis -> Plot -> ROI should give you the average (intensities/pixel)
November 7, 2011 at 11:31 pm #1397Markus StoeckliModerator[b]ramisir wrote:[/b]
[quote]So you guys think that the intensities from ROI regions CAN be used as a base for quantification as they are not cut or scaled?[/quote]
I would hope so, as we did many studies using this function…OK, here’s what I just tested with BioMap 3.8.0.3: I loaded an image with intensities above 1E6. I right-clicked on the image and selected Properties. In there, I switched Display Mode to Voxel and then clicked on Done. Then I clicked on the Zoom button and selected an area where I could easily identify 4 distinct pixels. Exit of zoom-mode with right-clicking and Done. With Analysis-PixelScan I got the values of the 4 pixels. Once done, I draw a region over these 4 pixels and looked at the ROI Statistics. The displayed “Mean” value was exactly the average of the 4 pixels. So this seems to work fine. Please describe in detail what you were testing.
Markus
November 7, 2011 at 11:47 pm #1398Dodge BaluyaParticipantI also did some testing using dummy data (see excel to analyze topic by markus for more details) in img form. The displayed average in the ROI panel is correct.
November 8, 2011 at 12:02 am #1399Markus StoeckliModerator[b]dodge wrote:[/b]
[quote]I also did some testing using dummy data (see excel to analyze topic by markus for more details) in img form. The displayed average in the ROI panel is correct.[/quote]I’m relieved, thanks B)
November 8, 2011 at 12:04 am #1400Ramin VismehParticipantHi Guys,
I did exactly the same thing as Markus said. That right! The number that we get from Analysis od ROI, is actualy the intensity of ions/pixels. So, it is nomalized to the number of pixels. I got the exact same numbers…
BUT….
What I am woried about is that in that 4 pixels which I chose, my ion of interest had an ion intensity of 32767! I am sure thats not the real intensity of my ion. for example if these 4 pixels that Markus talked about have all intensities above 32767, they all will be shown az 32767! I dont see any intensity higher than 32767 for any pixel when I do pixel scan. That means 32767 is also the limit for intensity!Markus:
I am curios to see what are the intensities of those pixels you chose? Arent they all 32767 or below?Ramin
November 8, 2011 at 12:37 am #1401Markus StoeckliModerator[b]ramisir wrote:[/b]
[quote]What I am woried about is that in that 4 pixels which I chose, my ion of interest had an ion intensity of 32767! I am sure thats not the real intensity of my ion.[/quote]Yes, there is an issue, but in the data file and not in BioMap. What instrument and software (version) do you use to acquire your data?
November 8, 2011 at 12:42 am #1402Ramin VismehParticipantI am using X-Calibur (version 2.1.0 1139) from thermo. The instrument is Orbitrap XL and I am doing DESI imaging! My BioMap is also 3.8.0.3.
November 8, 2011 at 12:51 am #1403Markus StoeckliModerator[b]ramisir wrote:[/b]
[quote]I am using X-Calibur (version 2.1.0 1139) from thermo. The instrument is Orbitrap XL and I am doing DESI imaging! My BioMap is also 3.8.0.3.[/quote]Hmm, I’m not familiar with the Thermo imaging implementation. I suggest that you contact Kerstin Strupat at Thermo in Bremen. She’s an expert in MS imaging on their instruments. Please let us know what you find out…
Markus
November 8, 2011 at 12:55 am #1404Ramin VismehParticipantAll right Markus…
Thanks. I will also check with Prosolia since I am using their Firefly software for coverting the raw data to image files. I will let you guys know the results. But you dont think that the intensities are being cut if they are above 32767 right? -
AuthorPosts
- You must be logged in to reply to this topic.