This story originally ran on June 7.
With AB Sciex's launch last month of its new TripleTOF 5600+ instrument featuring its Swath data-independent acquisition technique, interest in data-independent mass spec and, in particular, its use for targeted applications, is on the rise.
And while Swath â€" which the company developed in collaboration with Swiss Federal Institute of Technology Zurich researcher Ruedi Aebersold (PM 6/10/2011) â€" is perhaps the most fully developed such product currently offered by a major vendor, other instrument firms are pursuing their own DIA solutions, several of them in collaboration with University of Washington researcher Michael MacCoss, whose lab's open-source Skyline software system has emerged as a major platform for DIA data analysis and method development.
Traditionally, shotgun proteomics experiments used data-dependent acquisition wherein the mass spectrometer performed an initial scan of precursor ions entering the instrument and selected a sampling of those ions for fragmentation and generation of MS/MS spectra. Because instruments can't scan quickly enough to acquire all the precursors entering at a given moment, however, many ions â€" particularly low-abundance ions â€" are never selected for MS/MS fragmentation and so are not detected.
In DIA, on the other hand, the mass spec selects broad m/z windows and fragments all precursors in that window, allowing the machine to collect MS/MS spectra on all ions in a sample.
Enabled by the increasing scan speed of linear ion-trap instruments, the technique was introduced in a 2004 Nature Methods paper by Scripps Institute researcher John Yates III â€" who implemented it on a ThermoElectron LTQ instrument. Since then, DIA techniques have been employed by a number of proteomics researchers using a variety of different platforms.
What makes AB Sciex's Swath method interesting, MacCoss told ProteoMonitor, is its focus on targeted analysis of data. DIA has been around for nearly a decade, he said. "What's different about what people are doing now is that they are saying, 'OK, for this peptide I can predict the mass; I can predict the fragment ion masses; can I now find it in the [DIA] data?'"
"What AB Sciex has done [with Swath] is that they were the first people to kind of push this [targeted] workflow," he said.
Other vendors â€" particularly Waters, which introduced its MSE method in 2006 â€" have been marketing DIA technologies prior to AB Sciex's Swath launch, but, MacCoss said, vendors haven't really "pushed a solution for looking at analytes in the data in a targeted way [like] AB Sciex" now has.
"The data acquisition that is occurring now is no different than what's been done since 2004," he said. "The only thing that's different is that more recently the data analysis has changed to: Can we find peptide X in this data? As opposed to, let's try to find signals and [make peptide IDs]."
This shift could prove a significant one, said Brad Gibson, director of the chemistry and mass spectrometry core facility at the Buck Institute and an early-access user of the TripleTOF 5600+.
Gibson has been using Swath since the beginning of the year and, he told ProteoMonitor in an interview last month at the American Society for Mass Spectrometry annual meeting, in that time roughly 60 percent of his lab's experimental methods have come to incorporate the technique.
"You used to do a lot of data-dependent acquisitions," he said. "You would discover the landscape of your sample and what you wanted to follow up and then maybe you would make [multiple-reaction monitoring] assays, some kind of second, more focused set of experiments."
With Swath, though, "you can basically do discovery and targeted quantitation all in the same experiment," he said.
Such qual-quan workflows have been a major talking point for mass spec vendors for several years (PM 5/28/2010), but, Gibson suggested, with the release of the TripleTOF 5600+ and Swath, AB Sciex has for the moment taken the lead in terms of implementing them.
"I think AB Sciex may have the lead in terms of having [in the TripleTOF 5600] an instrument that is better suited to taking that kind of acquisition," he said, noting that one key aspect was that the instrument demonstrated very uniform levels of ion transmission across the breadth of its acquisition windows.
This uniformity, MacCoss said, is in large part a matter of optimizing and tuning the quadrupole used in the ion selection.
"Collecting a wide [acquisition] window is possible on any vendors' instruments," he said. "The difference comes down to whether or not there have been certain optimizations. AB Sciex has spent a lot of time trying to make it so that if you have a 25 m/z window [as is used in the Swath technique] it's pretty even in the transmission of signal across that window â€" so it doesn't matter if [an ion] is in the center of that window or at the edge of that window."
Again, though, MacCoss noted, more than the instrument itself, it's the emphasis on targeted data analysis and AB Sciex's software workflow for this analysis that sets Swath apart from other companies' offerings. And other mass spec vendors, both he and Gibson suggested, are focused on catching up â€" many of them in collaboration with MacCoss and his team's Skyline software tool.
Targeted analysis of DIA data works essentially the same way as the selected-reaction monitoring analyses for which Skyline was originally designed. This, MacCoss noted, has made the system a natural fit for such work, and, as mass spec vendors have scrambled to put together solutions for targeted DIA analysis, many have come to work with his lab.
"Mass spectrometry vendors want to sell mass spectrometers, but they know that unless they have a computational solution to process the data, they're going to have a hard time selling the instruments," he said. "So if they are behind [with regard to tools for targeted DIA work], what we've found is they tend to like to work with us because they can get a good solution quickly."
MacCoss and his colleagues have developed Skyline to support DIA methods from a number of vendors including AB Sciex, Waters, Agilent, and Thermo Fisher Scientific.
"I think most vendors have been aiding us because I think we have a solution that is as good as AB Sciex's [Swath]," he said.
MacCoss acknowledged that he was "of course, biased" with regard to that assessment, but Gibson â€" a sometime collaborator with the Skyline group â€" likewise suggested that the UW team was keeping pace, if not outpacing, its commercial counterparts.
"They're pushing the boundaries of trying to get the kind of software tools in an open source platform for doing these kinds of analyses," he said. "Every vendor is scrambling to do their own in-house versions, because they pretty much have to, but right now you see the open source community right in line with what the vendors are developing."
"And this is something very new," he added. "I've never seen it quite like this before where there's almost no lag time, where, in fact, you might even say that the open source community is even ahead of what the vendors have. The vendors are producing the instruments that allow you to do this, but already the open source and academic community is moving ahead in terms of how to do the data analysis and processing."
MacCoss suggested this is perhaps because his team features a number of world-class software engineers, including lead developer Brendan MacClean, who was formerly the lead developer on MicroSoft's Visual C++/Developer Studio project.
"Most academic labs [just] have some graduate students and post-docs writing software for their groups," he said. "The [Skyline team's] level of software engineering is not what you would typically consider academic software development."
In addition to their work on analysis of conventional DIA experiments, MacCoss and his team are developing tools for a new form of DIA called multiplexed DIA that, he said, can currently only be done on Thermo Scientific's Q Exactive instrument. The technique takes advantage of that instrument's combination of a quadrupole, a C-trap, and an Orbitrap to collect data in 20 m/z chunks, each composed of five separate 4 m/z windows. This, MacCoss said, enables improved performance by combining the sampling frequency provided by large acquisition windows with the specificity and reduced ion interference of smaller windows.
Gibson said he expects to see additional DIA innovations as use of the technique grows.
"Instrument vendors appear to be rapidly adopting [the technique], trying to figure out how they can get on the bandwagon," he said. "Is it going to be transformative? Is [targeted DIA analysis] something that is going to be a workflow that is going to be dominant in the next two years? I would place my bets that it probably will."