What is SWATH® Variable Window Acquisition, and why does it improve results quality?
To continue to improve the quality of the SWATH® Acquisition data that one can achieve in complex matrices, we are working on ways that we can decrease the size of the Q1 window, while still maintaining full mass range coverage and optimal cycle times. One great way to do this is to use the new Variable Window Acquisition feature that will be available in Analyst TF Software 1.7. To achieve better specificity in complex matrices, smaller Q1 windows (red line) are used in the m/z dense regions where many peptide precursors are measured, wider windows are used where there are fewer precursors, such that the entire mass range is still interrogated. The m/z density histograms (blue line) constructed from the TOF MS data for the proteome of interest (blue line) can be used to construct variable sized windows, where the density of precursors in each of the isolation windows is equalized across the m/z range.
How does the SWATH® Acquisition MicroApp select peptides and transitions for use in quantitation, when importing from a ProteinPilot Group file?
During the protein grouping process in ProteinPilot Software, an extensive analysis is performed on the proteins and peptides that are identified. These results are annotated according to which peptides are shared between proteins and which spectrum best represents the peptide. This information can be extracted by SWATH® Acquisition MicroApp during library extraction, along with other information that is used for peptide and fragment selection. Within the MicroApp, the user defines a range of filter criteria including peptide confidence, exclusion shared or modified peptides, and the number of peptides per protein as well as fragments per peptide. From the list of group file peptides that have not been excluded based on these filters, the peptides for extraction are selected from highest to lowest MS1 intensity.
To determine which fragment ions to use for each peptide, we assign three tiers of fragments for selection. The first tier comprises the fragment ions above the Q1 selection window, and ions are chosen from high to low intensity. If more ions are needed based on the user-defined number of fragments per peptide, the next tier of ions selected are from are the ions below the Q1 selection window and above the y3/b3 ion – again, from high to low intensity. That last tier of fragment ions is taken from within the Q1 selection window. Currently only y and b ions are used for extraction.
What software do I need to process my SWATH® Acquisition data?
If you are processing your SWATH® acquisition data using the SCIEX pipeline, there are three software packages that work together seam-lessly for processing. First,you wouldprocess your IDA data using ProteinPilot Software to create anion library.
Ion libraries only need to be created once and are loaded into the SWATH® Acquisition MicroApp in PeakView Software, which performs the extraction of XICs and determination of peak areas for the entire sample set. These peptide and protein peak areas then can be passed automatically to MarkerView Software that provides visualization and statistical analysis tools for assessing differential expression information. Other search engines can be used to create ion libraries, and then the information can be imported into the SWATH® Acquisition MicroApp using the mzIdentML format. Peptide and protein peak areas can also be exported from the software in other formats to integrate with other downstream tools. In addition to SCIEX tools, there are a growing number of publicly available software solutions that can also be used to process SWATH® acquisition data.
Is it possible to create a theoretical ion library?
In the current data processing tools,the algorithm for peak group detection compares the pattern of intensities of the fragment ion peak areas from the SWATH® Acquisition data with theintensities of the peptide MS/MS, as obtained from previous MS analysis. A good pattern match is one of the key elements to the scoring of the peak groups. Retention time of peptide elution is also a key element.
So if a theoretical library was to be generated, these are two elements that would need to be predicted for good performance during data processing. There are a few good algorithms for retention prediction that would work well for this purpose. However, prediction of the relative pattern of fragment ion has been a more difficult problem to date. If this could be done well, then theoretical library generation becomes more feasible.
What is the best strategy for performing retention time calibration when using SWATH® Acquisition?
When processing SWATH® Acquisition data, you want to use a reasonably narrow time window to reduce the chance of incorrect peak integrations. There are currently two strategies being employed today to adjust for differences between the retention times in the spectral ion libraries and the retention times observed in the current sample analysis. Some labs add standard peptides to their samples, both when creating the ion libraries and when running SWATH® Acquisition. This enables the retention times for targeted extractions to be adjusted to correlate with the current acquisition. Retention time alignment can also be done using some of the medium-to-high abundance endogenous peptides present in both the ion library and the current sample. In each case, you want to have confident detections across the whole gradient in order to adjust the retention time scale for the entire sample.
SWATH® Acquisition looks like an exciting technology for targeted quantitative proteomics, but how does it compare to the industry standard MRM technology for quantitative profiling?
Of course, this will depend on which triple quadrupole or QTRAP system you are comparing to SWATH® Acquisition. We did some work to compare the top end QTRAP 6500 system with SWATH® Acquisition on the TripleTOF system to understand their relative performance in quantitative profiling. With SWATH® Acquisition, you can quantify very large numbers of peptides and proteins from a single acquisition and achieve quantitative reproducibility that approaches that of MRM analysis. On the QTRAP 6500 system, the numbers of MRMs that can be scheduled are far fewer than that of SWATH® Acquisition, but due to the higher sensitivity of the QTRAP 6500 system and the specificity of MRM, we were able to get better quantitative data on the very low abundant peptides in the samples. One key point is that it is very straightforward to transition from SWATH® Acquisition to MRM on these platforms, since they use the same ion source and collision cell technology; accordingly, all of the assay information can directly transfer.
What are the critical acquisition attributes of a mass spectrometer required for good SWATH® Acquisition data quality?
Indata-independentacquisition strategies like SWATH®, an expanded mass isolation window is stepped across a mass range covering the mass-to-charge (m/z) distribution of peptides, and a full scan MS/MS spectrum is collected at each step. High resolution MS/MS is one key attribute, as it enables target fragment ions to be found and extracted with specificity from these complex MS/MS spectra. More importantly, these MS/MS spectra must be acquired at high speed. We find that narrower isolation windows yield higher specificity in data analysis, which means that it is advantageous to take more steps to cover the mass range of your experiment and collect the MS/MS faster to maintain an optimal total cycle time. This is why the TripleTOF systems are so powerful for this acquisition technique: because high resolution MS/MS can be acquired at very high speed, much smaller isolation windows can be used.
What is the best way to build spectral ion libraries for SWATH® Acquisition data processing?
The information needed for targeted data extraction in our current SWATH® workflow is relatively simple. You need the parent ion m/z, the m/z and relative intensities of the major fragment ions that are produced during MS/MS, and you need to know the relative retention times of the peptides of interest. Currently, researchers are employing two strategies: a data-driven approach or a repository approach. In the data-driven approach, extensive data- or information-dependent analysis (IDA) is performed on the sample of interest, and proteins and peptides are identified from this acquisition using ProteinPilot or other database search tools. The identified peptides in the search result then serve as an ion library to apply to the SWATH® acquisition runs. This interrogation can be accomplished with 1D LCMS or 2D LCMS depending on the depth of library required for the study. In the informatics approach, MS/MS data from public or internal repositories are mined for target proteins and peptides, and are used to create the ion library. Here, retention time calibration strategies should be employed.