Chromatogram Subtraction

[stltestlab]

Good day.

I was wondering if anyone knows of a way to subtract an entire chromatogram from another chromatogram.

For example, I am running a painted polycarbonate sample. I have isolated the paint and polycarb and have two separate chromatograms.

Now I would like to subtract the polycarb chromatogram from the paint chromatogram if is is possible in Chemstation.

Anyone know if it is possible to do this type of subtraction?

Thanks.

[dblux_]

Good day.

I was wondering if anyone knows of a way to subtract an entire chromatogram from another chromatogram.

For example, I am running a painted polycarbonate sample. I have isolated the paint and polycarb and have two separate chromatograms.

Now I would like to subtract the polycarb chromatogram from the paint chromatogram if is is possible in Chemstation.

Anyone know if it is possible to do this type of subtraction?

Thanks.

Naturaly it is possible to subtract one chromatography signal from another in ChemStation.


Bearing in mind definition of chromatogram I discourage calling it "chromatogram subtraction".

[chemstation]

hi stltestlab

Are you looking for trace chemicals that are in the paint but not in the polycarbonate, or TIC profiling

If for trace chemicals check out viewtopic.php?f=4&t=9207
I wrote about using AMDIS to achieve this.

If for TIC, have you tried the overlay functionality ?

I did write a programme that did exactly that, however because I was interested in trace analysis, the very
slight shifts in retention times caused artificial TIC peaks https://sites.google.com/site/chemstation/
which is why AMDIS is better, as it creates a list of compounds, that you can then put in a library (as background) and then search against you paint sample, and any peaks without a library match, must be unique to the sample.

Alex

[lmh]

you're teetering on the brink of multivariate analysis. A very important issue is how to handle retention time drift. If you merely subtract a chromatogram from another, if a peak has shifted very slightly, you'll get a strange narrow positive peak at one side, and a negative peak at the other. Exactly what the "right" thing is to do depends on what question you're trying to answer... For a quick look-see, the overlay option is the easiest. Chemstation has a toolbar allowing overlaid chromatograms. Someone mentioned TICs. If you're doing LC-MS you are definitely in the realms of multivariate analysis and could consider preprocessing by packages such as XCMS, which will at least reduce your data-set to an aligned peak-list.