Ananlysis of MS/MS data from Waters Instrument

[Abhishek Dubey]

Hi all

I am acquiring my MS/MS data on Waters Instrument. The way I do it is first take a MS1 spectrum, then  choose the peak to be fragmented and then fragment it. For each fragmentation spectra there are multiple scans. For fragmentation of one peak, one file is generated. 

I wanted to know before analyzing in crux, do you combine the fragmented spectra ( MS/MS ) of different peaks observed in MS1 spectra or do you analyze one peak at a time in crux ?

[Will Fondrie]

Hi Abhishek,

It sounds like you've performed a MALDI-MS/MS experiment. Is this correct?

If it is a MALDI-MS/MS experiment, then summing MS/MS scans originating from the same precursor is appropriate. However, I would not combine MS/MS scans from different precursors. I'm fairly sure the vendor software has a way to do this automatically, but you'll have check with them to find the appropriate option.

The side effect of leaving the scans separate is that you may have many peptide-spectrum matches (PSMs) for the same precursor (one for each scan), but each will likely be of lower quality than if the scans were summed. Summing MS/MS scans for the same precursor on a time-of-flight instrument results in a higher signal-to-noise ratio and a more confident PSM.

Best,

Will

[Abhishek Dubey]

On Saturday, March 7, 2020 at 12:01:55 AM UTC+5:30, Will Fondrie wrote:

Hi Abhishek,

It sounds like you've performed a MALDI-MS/MS experiment. Is this correct?

Yes I performed MALDI-MS/MS experiment.

 

If it is a MALDI-MS/MS experiment, then summing MS/MS scans originating from the same precursor is appropriate. However, I would not combine MS/MS scans from different precursors. I'm fairly sure the vendor software has a way to do this automatically, but you'll have check with them to find the appropriate option.

The side effect of leaving the scans separate is that you may have many peptide-spectrum matches (PSMs) for the same precursor (one for each scan), but each will likely be of lower quality than if the scans were summed. Summing MS/MS scans for the same precursor on a time-of-flight instrument results in a higher signal-to-noise ratio and a more confident PSM.

Thank you for your reply. The file generated contains different scans of same precursor ion fragmented at different energies. However, when I am analyzing I note that lnNUMdsp value is one same value for all target psms and another same value for all decoy psms.

One more doubt is that indiviual scan also consist the precursor ion peak. Is it ok to have that peak while analyzing the data ? 

[Will Fondrie]

Hi Abhishek,

Thank you for the details of your experiment.

The file generated contains different scans of same precursor ion fragmented at different energies.

In this case, I think it would be best not to combine scans. Typically this is done when all scans are collected under similar conditions. However, if you do have multiple scans using the same fragmentation energy, I would still advise combining those.

However, when I am analyzing I note that lnNUMdsp value is one same value for all target psms and another same value for all decoy psms.

Is this the same issue that is referenced in this thread?  https://groups.google.com/d/topic/crux-users/YZ-oxH6NQIw/discussion

One more doubt is that indiviual scan also consist the precursor ion peak. Is it ok to have that peak while analyzing the data ? 

As I understand it, your concern is that some MS/MS scans contain unfragmented precursor ions, is that correct? If this is the case, you can certainly still analyze these. Tide-search has the "--remove-precursor-peak" that can be used if you are concerned about the unfragmented precursor ion peak affecting your results.




Best,

Will