Reading Intensities Correctly
zrolfs opened this issue · 6 comments
Hello! Thank you for this excellent repository. I'm trying to use read_chemstation_ch.R to read in a v179 .ch GC-FID file. The retention times seem right, but the intensities are very different from what I see when I open the file in MassHunter. Could there be an issue with the scaling factor that is applied?
I've attached the .ch (FID1A.txt) file as well as an excel where I compared the MassHunter and read_chemstation_ch intensities. Any help is appreciated!
Hi zrolfs,
Thanks for pointing this out (and sharing your file)! There seem to be a couple of different issues at play here. The first is that there are at least two different types of chemstation 179 files. I was thinking that one was generated by openlab (and are accordingly marked "OL") while the other is generated by chemstation (and are marked "GC") but apparently this is not the case, which leaves me uncertain how to determine programmatically which type of 179 file is at hand. It seems that your file is the less common type, at least amongst the small subset of files i've been able to collect so far. I noticed that your file says "Mustang Chemstation" in the head (as does another one of my files which appears to (maybe?) be in the same format), while most of the other 179 files I have are from "Asterix Chemstation". I'm wondering if this could be a meaningful distinction as far as what type of 179 files are being produced.
The second issue is with the intensities. The intensities you plotted from masshunter appear to be uncorrected intensities (e.g. the uncorrected intensity at 4.49 minutes is 171572.7, compared to the 172000 you got from masshunter), but according to my understanding there's a scaling factor encoded in the file at byte 4732 (in this case 0.00013), so I'm not sure why the unscaled intensities are showing up in masshunter. I presume the data were originally collected in Chemstation. I don't suppose you would by any chance have access to the original Chemstation workstation to compare the values with the ones showing up in masshunter?
Here's a figure showing the uncorrected intensities:
and corrected intensities:
Ethan
So I went ahead and opened your file in GC Chemstation (Rev. B.04.02 SP1 [212]) and it looks like the scaled version:
So I think MassHunter is wrong. Either that or there's a way to toggle the scaling of the y-axis and you may have it turned off.
If so, the only issue is to figure out how to determine programmatically which version a 179 file belongs to. I'm thinking of just going with my hypothesis that the "mustang chemstation" produces your style of file, but I don't really know if that's correct
Hi @zrolfs. I tried to fix this on the dev branch by telling the parser to assume the "mustang" files are in the 8-byte format (like your file). Want to test it out for me? You should be able to install it by running remotes::install_github("ethanbass/chromConverter", ref="dev"). You'll likely need to reload your R session to purge the old version from your cache.
P.S.
Would you mind if I included your file in my chromConverterExtraTests package? It would be helpful to have a test file with your 8-byte format. I would be happy to attribute you of course.
Hi Ethan,
Thanks for the quick response and your work on this! My wife gave birth between our messages, so I am going to be MIA for a bit. I should be back on January 2nd, but I've reached out to colleagues that might take a look at this for me in the meantime.
Quick thoughts:
-I agree with your finding; I think MassHunter is wrong, but at least proportional.
-Feel free to include the file; no attribution needed :)
Congratulations!!!! hope you and the family have a nice holiday