Skip to main content
Premium Trial:

Request an Annual Quote

Top-Down Proteomics Software Package Aims to Ease Analysis of Ion Mobility Data


NEW YORK (GenomeWeb) – University of Michigan researchers have developed new software tools for ion mobility mass spec-based top-down proteomics.

Described in a paper published last month in Analytical Chemistry, the tools allow scientists to process top-down ion mobility data in an automated way. This could significantly ease the processing and interpretation of this data and potentially help make ion mobility gain wider use among top-down mass spec researchers, said Alexey Nesvizhskii, professor of computational medicine and bioinformatics at Michigan and senior author of the paper.

Nesvizhskii said that he developed the software tools, called IMTBX and Grppr, as part of an effort that was initially focused on traditional bottom-up proteomics. But he was drawn to look at top-down proteomics by his Michigan colleague Brandon Ruotolo, whose lab works on ion mobility mass spec approaches to study intact proteins and protein complexes.

"Our interest was initially more in developing a robust computational tool for ion mobility data in general," Nesvizhskii said. "But from talking to [Ruotolo's lab], we realized that they have a really great need for this kind of software. So we changed our focus and started working with them to demonstrate how our computational methods could automate top-down analysis."

Unlike conventional shotgun proteomics, where proteins are digested into smaller peptides prior to mass spec analysis, top-down proteomics looks at undigested proteins. Analyzing intact proteins could potentially allow researchers to better characterize protein isoforms and post-translational modifications that can be lost or difficult to analyze using traditional bottom-up methods.

Top-down is more technically challenging than bottom-up methods, however, and researchers are still working out many aspects of the workflow necessary to allow for broad, sensitive, and high-throughput characterization of intact proteins in complex biological samples.

FTICR and Orbitrap instruments are most commonly used for top-down work, Nesvizhskii noted, due to their higher resolution compared to TOF instruments.

Spectra generated in a top-down experiment are highly complex, and "it is really hard to identify accurately different isotopes and different fragments in this type of data," he said. The high mass accuracy and resolution of FTICR and Orbitrap instruments make them well-suited to analyzing such data.

However, Nesvizhskii said, these instruments "are pretty slow in the sense that to acquire enough signal on those highly charged fragments with high mass resolution, you have to accumulate ions for a long time." This leads to a low duty cycle, which, the authors wrote, compounds "the challenge of low [signal-to-noise ratio] for fragment ions, as fewer spectra can be acquired to average and improve spectral quality."

Top-down researchers are working to address these limitations and have made significant strides using Orbitrap instruments. Use of QTOF instruments coupled to ion mobility could make for an interesting alternative approach, though, Nesvizhskii said.

Ion mobility uses differences in size, shape, and charge to separate ions in the gas phase. Mass spec vendors including Waters, Agilent, Sciex, and Bruker offer instruments incorporating IMS devices, which, in proteomics, researchers typically use to provide an additional layer of separation after conventional LC.

"If you can couple ion mobility with a TOF instrument, you can get improved resolution from the ion mobility, but the sequencing speed will still be high because the TOF instruments are fast, so you can get improved speed and still have the [necessary] resolution," Nesvizhskii said. "That is where the promise is and why there is interest in using ion mobility for top down."

However, ion mobility data, and top-down ion mobility data in particular, is complicated and noisy, he added. "You have to have some good signal processing to get rid of the noise and find the real peaks in the data."

A number of software packages for automated processing of top-down mass spec data exist, but they are not well-suited to handling ion mobility mass spec (IM-MS) data.

Nesvizhskii said that Ruotolo's lab has developed semi-automated techniques for processing and analyzing its top-down ion mobility data, but that the approach is still too labor intensive to win wide adoption.

As the authors wrote, "all reports of IM-MS top-down to date, to our knowledge, rely largely or entirely on manual interpretation of spectra. The complexity of fragment ion populations generated in top-down experiments makes this a massively time-consuming process, as hundreds to thousands of peaks must be identified in each dataset, severely limiting the scope and speed of viable experiments."

One of the main challenges, Nesvizhskii said, was working with the raw data generated by the Waters instrument used in the study, a Synapt G2 HDMS IM-QTOF.

"There was no easy way to extract [raw] data from the Waters instrument," he said, noting that, compared to Orbitrap instruments, there are considerably fewer open-source computational tools capable of working with raw data from Waters instruments. "Waters helped us quite significantly in being able to read the raw data."

The IMTBX and Grppr software package processes the raw top-down IM-MS fragmentation data to automatically generate a monoisotopic peak list that can then be analyzed using various sequencing tools.

While the study used data generated on a Waters instrument, Nesvizhskii said he is interested in applying the software to top-down IM-MS data from other vendors who offer ion mobility coupled to TOF mass specs. He cited Bruker as a vendor whose instruments he and his colleagues are interested in looking at next, with Agilent being a possibility, as well. Sciex also offers TOF instruments and ion mobility.

While all of these vendors use somewhat different ion mobility approaches, Nesvizhskii said he believed the tools developed by his group would be broadly applicable.

"There are some differences in the principles behind how [the different ion mobility systems] work, but we looked into it and fundamentally, the data look very similar," he said. "It all comes down to our ability to read the raw files, but once we have the ability to do that, there are only minimal changes necessary in the algorithms to support the processing of data from [other vendors'] instruments. We hope that in the future, our toolbox will support all instruments that generate ion mobility data."

He and his team are also continuing their work on developing the software to improve analysis of IM-MS data from conventional bottom-up experiments. He suggested that ion mobility is less widely used in proteomics than it might otherwise be due to a lack of open-source informatics tools for analyzing this data.

"Even though the technology has been around for a while, the computation tools are less mature for ion mobility-based technology than for more conventional applications," he said. "So we want to develop tools for ion mobility data."