Current proteomics instrumentation is perhaps best analogized to the automobiles of 80 years ago, Bruno Domon, head of the Luxembourg Clinical Proteomics Center, told ProteoMonitor this week, describing efforts afoot within the field to simplify and make more accessible the tools of the trade.
"If you look at the car industry in Europe in the 1930s, basically, these were very expensive machines that were reserved for an elite and required experts for maintenance and driving," he said.
And so it is today, Domon suggested, with mass spec-based proteomics. While the field continues to draw interest from a growing number of non-specialists hoping to apply its techniques to their research questions, many others fail to take advantage of its potential due to the difficulty and expense involved.
"If you look back at what biologists were doing [for protein measurements] 20 years ago, it was Western blots," Domon said. "And if you look today, it is still largely Western blots. So the question is, 'Can we do something more modern?'"
Domon also noted that proteomics researchers have watched the evolution of genomic techniques over the last decade as analysis of whole genomes has become relatively automated and high-throughput. Proteomics, on the other hand, "is still struggling to do [whole proteome analyses] in a routine way," he said.
This state of affairs was the motivation for a session on making proteomics instrumentation more accessible that Domon led last week at the Human Proteome Organization's 12th annual meeting in Yokohama, Japan (PM 9/20/2013).
The session, which Domon said emerged out of less formal discussions between researchers at the 10th HUPO in Geneva, aimed to outline requirements for more broadly accessible mass spec devices and sketch out potential routes for achieving them.
The matter was split among two working groups: one focused on an instrument for targeted protein measurements – the sort of work commonly done by Western blots – and the other focused on an instrument for shotgun-style proteome profiling, which the organizers analogized to RNA-seq.
For the targeted device, the session participants determined that such an instrument would ideally offer around six or seven orders of dynamic range while allowing users to target tens to hundreds of proteins per experiment, said Mark Cafazzo, AB Sciex's global market manager for proteomics, who co-chaired this working group along with Swiss Federal Institute of Technology Zurich researcher Ruedi Aebersold.
On the shotgun side, the participants called for an instrument capable of identifying and semi-quantifying every protein in a cell at an average of roughly 50 percent sequence coverage and reproducibly measuring differences between two cell populations, said Andreas Huhmer, proteomics marketing director at Thermo Fisher and chair of that working group.
In both cases, the agreed upon requirements are either achievable or close to achievable from a technical standpoint, Cafazzo and Huhmer told ProteoMonitor. The real challenge, they said, lies in packaging these capabilities in platforms easy enough for non-experts to use and inexpensive enough for them to afford.
For instance, Huhmer said, in studies over the last few years, scientists including Aebersold, Utrecht University researcher Albert Heck, and Max Planck Institute researcher Matthias Mann have managed to identify essentially the entire proteome accessible via standard LC-MS/MS workflows.
However, these groups have "optimized every step in that workflow" from sample preparation through to data analysis, he noted. "They have years and years of experience, and so you would have to essentially pack all that experience into a product if you were to build something like" the shotgun instrument described in the working group.
Likewise with the targeted device, Domon said. "You have a biological sample and what you want is an answer," he noted. "In between you have sample prep; you may have separations; you have mass measurement; and then you have data processing to eventually give you an answer that is useful to a biologist – for instance, this biomarker is present in blood at a concentration of 53 nanograms per mL. So we're talking about an integrated solution."
Cafazzo said that given this need, he expected that, ultimately, trade-offs between performance and accessibility would be resolved in favor of the latter.
"My assessment is that ease of use and multiplexing are the most important things, and then the question is, 'How many orders of dynamic range can you give us? How much sensitivity can you give us?' With the caveat being that it has to be easy to use and have quick turnaround," he said.
More than identifying specific instrument designs, the purpose of the HUPO session was to "identify the needs so that we can give [the requirements] to our research team to address in one way or another," Cafazzo said.
For instance, while "a lot of the requirements [put forth for the targeted platform] certainly sound like some sort of automated LC-MS tool using a specific set of [multiple-reaction monitoring] assays, who is to say that's really the best way to address that problem?" he said. "We need to let out creative thinkers in research develop something that addresses the market need more than bringing them some prescribed solution."
For mass spec vendors, determining what exactly that market need is will be a key part of the effort. On the targeted side of things, AB Sciex has done research suggesting that the current market for Western blots in proteomics research exceeds $200 million.
"Of course, this number doesn’t represent the entire Western blot market," Caffazo said, "but it represents a segment that could be immediately addressable with a mass spectrometry-based analyzer."
Huhmer declined to project the market for the shotgun-style analyzer, saying that this would merely be "speculation" on his part. He said that part of the HUPO effort going forward would be "to figure out whether biologists want to do these types of experiments, and at what price."
This, Cafazzo noted, means that, though the conversation began at HUPO, it will need to be taken to the broader biology community. "A room full of proteomics people is not necessarily who we need to be talking to right now to define these [systems]," he said.
To obtain input from outside the proteomics community, the session organizers are creating a discussion forum on the HUPO website within which they intend to build a survey on the matter aimed at general biologists, Cafazzo said.
He added that while another proteome analyzer session was slated for next year's HUPO meeting, in the meantime the project would likely be carried forward most effectively by the organization's Industrial Advisory Board – a collection of proteomics vendors that meets on a monthly basis.
After all, Cafazzo noted, "it really comes down to the vendors to decide whether or not these types of analyzers fit in with our business strategy and whether we see a market for this type of solution."
As for current level of vendor interest, Cafazzo said that, in the case of AB Sciex, the company is "actively investigating what the market opportunity is and what our capabilities are to bring an effective solution to market right now."
"We are always looking for whatever that next opportunity is and it really comes down to what capabilities we have in our research pipeline and our development pipeline and whether or not we can bring something to market that addresses the majority of these needs that have been identified," he said.