Derek-
Well it looks like bill mahn has you taken care of. I'd add a few things that may not
solve all the issues as well as bill's suggestions, but may supplement them.
Matt Sweeney
mattsweeney@earthlink.net
Mass Spec Consulting
Training/Operations/Consulting/Method Development
LC/MS Pharmacokinetics, Peptides, Proteins, Metabolism,
Maintenance Classes, Specialist in Finnigan Equipment and Software
Firstly, I've been using the
>Zoomscan feature to check the charge state of a number of tryptic peptides
>and have been having difficulties getting interpretable results as the
>signal flicks between +1 and +2 readings sometimes going as high as +3. The
>signals I am looking at aren't always the strongest in the spectrum (NL
>approx 10e-3/-2) and usually vary quite markedly in their intensity.
Yes this may happen at those count levels due to statistics of ion counting. Not much to
do except get more signal or signal average. If you are not averaging then you are not
getting the data that you could be getting. I set the averager to 100 or so and then turn
it on when I am sitting on an ion and let it go until I can assign the charge state. It
can take a while but it does help.
Some ions just plain don't zoom very well. I think they fall apart in the trap. I have
seen this on some glycopeptides and other species. If the species is a non-covalent
species (solvent adducted or something) then it may be too delicate to zoom scan. You
know them when you get them.
All ions are not created equal! If you are in the 10^2s then that is NOTHING...well 100
ions or so(actualy more but that is what the scale "says"). They may be real and chemical
noise overlayed on each other and the chemical noise may be more delicate and cause moving
signal over your "real" ions.
>Subsequently when I attempt MS/MS on the signal, however much I manipulate
>the isolation width (generally between 1.0-3.0)
Don't bother with 1.0 use 3, 4 or 5. I often start my MS/MS by first isolating the ion
with basically 0 rel energy. Then I play with the isolation width to see how much I loose
by cutting it down to 1.0, I usually start at 5.0 and either leave it there or cut it to
3. Then I bump up the RE to say 10 or 20 and see what happens.
> I always seem to get some
>'stray' signals either side of the main signal. For instance I was looking
>at the fragmentation of a peptide of m/z 938.6 and the parent signal seemed
>to be coming from m/z 937.0 - 938.0.
See Bill's concern about calibration. Remeber that the at about 1000 there is a marked
difference between mono-isotopic and avg mass(avg is bigger). I like to really look at
the ion in the full scan up close (zoom the display not the scan) before I choose.
>Is that sort of variation
>normal/acceptable?
Well it CAN be depending on what you do. I'd rather not have it however. If you just spit
the scans to sequest/mascot then the parent mass tolerance can be set to still get a hit
using the ms/ms data.
>>Secondly, having had some zoomscan results indicating a +1 charge state,
>when I perform the MS/MS on the same ion I still get peaks at higher m/z's,
>suggestive of a higher charge state. Are there any diagnostic tests I can
>perform to check the performance/calibration of our machine?
See Bill's comments. If there is an ion that is +1 it may be "hidding" and iso-baric ion
beneath of lower intensity that is +2/+3! That would cause this. You also get SOME
electrical noise in all MS detectors, ALL of them. That could occur any where in the mass
range. Then don't forget cosmic rays and x-rays. Neutrinos are not so likely. Also the
solvent and other vapors enter the trap and can interact with the "parent" ion during the
VERY long (seconds) trapping times that faint ions are forced to endure before being
scanned out (this is the ion-molecule reaction). These can be observed. If the high mass
signals repeat themselves then it is iso-baric species or an ion molecule product most
likely. Ion-molecule products that I have observed don't usually allow much of themselves
to be used as subjects of MS^n or zoom-scan. Some do I am sure but the ones I have
bothered to chase have not. I ignore them. The presence of real frag ions and a good
parent mass allow a good hit in DBase searches even in the presence of a few of these when
they do occur.
>
>Thanks and regards,
>
>Derek Bradley
>Department of Medicine
>UCL
>
>
This archive was generated by hypermail 2b29 : Wed Nov 22 2000 - 08:37:58 EST