Christie Hunter


Optimized rolling collision energy curves for IDA and SWATH DIA for peptides

During data dependent acquisition (DDA or IDA) or SWATH acquisition, the collision energy can be automatically adjusted according to the mass/charge and charge of the peptide. This dependency has been well characterized on our QTOF systems. By selecting rolling collision energy (Analyst software) or dynamic collision energy (SCIEX OS software) this function is invoked during acquisition.

These collision energy curves were developed from 1000s of peptides, by determining the collision energy that provided the most signal intensity in the higher m/z y-ions.  They were first developed using SWATH DIA on 3 different TripleTOF 6600 systems and then validated on additional instruments for both DDA and DIA acquisition. More recently these curves have been confirmed for use on the ZenoTOF 7600 system.

These are the defaults settings in SCIEX OS software when you select Dynamic collision energy.  If you are using Analyst software, please download this table and enter these values into the IDA CE Parameters Table found under the Scripts menu.


Join the discussion

Comment below on this article and our team will answer your questions.


  1. Richard Kay

    Hi Christie
    We have recently got a 7600 which we are using for peptidomics, which means that we are seeing some large and highly charged peptides in our samples. With regards to the dynamic collision energy assignment algorithm, does the software identify the charge state of peptide present within a given swath window and then apply the appropriate dynamic collision energy setting?

    Was the dynamic collision energy table assigned using a proteomics experiment? has one been set up or considered for peptidomics experiments? What about undefined charge states – would it be better to set this to a higher charge state peptide in our peptidomics experiments – for example default at 4+?

    • Christie Hunter

      The dynamic collision energy curves were developed using a proteome digest, so based on fragmentation of tryptic peptides. So if your peptides are non-tryptic or digested with a different digest agent, it is possible the curves would need to be adjusted a little bit. If you wanted to try playing with this, I would try first to just adjust the intercept, keeping the slope constant. Trying adjusting the slope by increments of say plus/minus 3 V and see how that impacts either the quality/number of your peptide IDs in DDA or your peptide IDs/quantitation in DIA.

      When performing DDA experiments, there is a peak detector that runs in real time to identify the charge state of the selected peak so the collision energy can be selected using the correct equation from the table. If the peak charge state is unknown, then the equation for unknown charge state is used. You can go in and edit that equation for your method in SCIEX OS software and set that to be more similar to a high charge state as you suggest if you think that peaks getting assigned an unknown charge (charge state = 0) are typically high charge rather than 1+. If you open your DDA file in IDA Explorer and look at the Table view, you can sort your Charge column and view any Charge =0 spectra to determine what type of peaks are not getting assigned.

      For SWATH DIA experiments, collision energy is not assigned by peak because a lot of different peptides of different charge states will be fragmented in each MS/MS. You define the collision energy equation to be used for the whole experiment by selecting the Charge state in your method, so select the charge state that reflects the majority of the species in your sample. Toggling this Charge state field will compute new collision energies across your SWATH DIA windows. Note if you paste in a new mass table, make sure you toggle the Charge state to force it to recompute the collision energies.

      If you want to build your own CE curves for your specific type of peptides, I would be happy to walk you through how I did these experiments to build these CE curves on a call, just reach out!

  2. Richard Kay

    Thanks Christie.
    We had already made some tables looking at how changing the rolling collision energy equation changed the energy for given m/z and charge state. We will look at the search output and look at the MS/MS spectra for specific peptides and see how well the system fragments highly charged peptides (looking at remaining precursor and how many suitable product ions are formed).

    Definitely some things to try going forward, thanks very much for the help!

  3. Richard Kay

    Hi Christie
    Me again… We are potentially looking at using a fixed CE with a CE spread, as we are dealing with peptides of varying m/z values and with some very high charge states. We did this on our Orbitrap system and it worked pretty well for us on that system.

    Trying to get an optimised rolling collision energy setting is proving a little challenging, as essentially each peptide will be selected to a single collision energy value. Using the CE spread function will allow us to get multiple different CE values applied to each peptide. This can then give us three throws of the dice to get an optimum CE value so to speak.

    If we select the CE spread function, can I check – the system will then spend 1/3 of the accumulation time at each of the three possible CE values for each MS/MS spectra? For example, setting CE at 40 +/_ 10 will mean 1/3 of time will b e30, 1/3 at 40 and 1/3 at 50?

    • Christie Hunter

      Glad you are making progress in optimizing your acquisition conditions!

      On the QTOF instruments from SCIEX, the CE spread is done with a larger series of steps across your specified range, so it is more of a ramp. When using CES in DDA or DIA mode, make sure you keep your accumulation times at 20 msec or above to ensure there is time for the instrument to get all the way through the ramp.

      Good luck!!

Submit a Comment

Wordpress Social Share Plugin powered by Ultimatelysocial