Vienna Scientists Set a New Standard for Investigating the Functions of our Microbiome

1.Beyond Who’s There and Why Function Matters

Knowing which microbes are in the gut isn’t the same as knowing what they do. Standard microbiome sequencing can tell you who’s there, but not which microorganisms are active, what they’re doing, or how they interact with each other and the host.

To close that gap, microbiome scientists use metaproteomics, which measures the actual proteins microbes produce. This reveals which functions are truly active, not just which ones are genetically possible. Metaproteomics relies on mass spectrometry, and one of the most widely used technologies in the field is PASEF (trapped ion mobility spectrometry with parallel accumulation-serial fragmentation).

PASEF has already proven itself in recent years through its greater taxonomic and functional resolution, higher throughput and lower detection limits. Besides classical mass separation, PASEF exploits an additional physical dimension: ion mobility, how quickly differently shaped molecules move through a gas. This extra sorting step makes it easier to tell real protein signals apart from background noise in a highly complex sample like human stool.

On this platform, several measurement strategies exist, differing in how they sample this ion mobility and mass space. Two are increasingly used as standard tools in microbiome research: data-dependent (DDA-PASEF) and data-independent (DIA-PASEF). DDA-PASEF selects individual, particularly promising signals and measures each one precisely, but as a result does not capture every signal. DIA-PASEF, on the other hand, systematically measures broad segments across the entire signal range to maximize completeness. This broad approach can make it harder to assign individual signals to a specific molecule with full certainty, though it does not come at the cost of overall measurement precision. More recently, several refined approaches, Slice-, Synchro-, and midia-PASEF, have promised further gains in identification depth, sensitivity, and reliability by combining this broad coverage with a more targeted way of capturing individual signals.

But how well do any of these methods actually hold up in highly complex metaproteomes like the gut microbiome, and is one variant better suited to a given research question than another?

That’s the question Feng Xian, David Gómez-Varela and Manuela Schmidt at the University of Vienna set out to answer in their new Nature Communications paper: which PASEF method performs best, and which one is reliably reproducible. The work came out of the Center of Excellence for Metaproteomics, a centre the University of Vienna set up together with instrument manufacturer Bruker Daltonics in 2024.

“Metaproteomics allows us to see what microbial communities are actually doing, rather than simply which microbes are present,” first author Feng Xian said in a statement released by the University of Vienna. “By identifying the most suitable analytical strategies, we hope to make future microbiome research more sensitive, reproducible, and accessible across medicine, environmental science, and biotechnology.”

2.Putting Five Methods to the Test

To test the efficacy of the different PASEF approaches, the researchers from Vienna spiked complex human stool samples, pooled from multiple donors, with two defined bacterial species, Ligilactobacillus murinus and Salinibacter ruber, added at three defined concentrations, alongside samples with no added bacteria as a control. Because the researchers measured proteins rather than species directly, this created a known ground truth for how reliably each method could trace detected peptides back to their species of origin, not just how much it detected overall, but how accurately. Each condition was also tested across three different run lengths: 5, 22, and 45 minutes. Shorter runs favor speed, while longer runs give the sample more time to separate before being measured, typically revealing more detail.

Based on the benchmarking, the authors were able to offer practical guidance for choosing the right method depending on the goal. If the priority is throughput, reproducibility, and the ability to scale up to large cohort studies, such as population-wide microbiome surveys or clinical studies comparing patients to healthy controls across hundreds or thousands of samples, DIA-PASEF is the best overall choice, offering a strong balance of identification depth and reliable quantification.

If the priority is sensitivity, for example when working with samples that contain very little biomass, such as environmental samples like soil or water, or where only a small amount of material can be collected, such as skin or tissue swabs, Slice-PASEF performs best, offering the greatest depth and functional detail, though this comes at the cost of larger data files and longer processing times compared to DIA-PASEF. In exploratory studies, such as building microbial reference databases from scratch or identifying previously uncharacterized proteins through de novo sequencing, DDA-PASEF remains the better choice. Synchro- and midia-PASEF gave more mixed results, performing well in some conditions but not others. Both are still seen as promising. Midia-PASEF is designed to assign signals to the right molecule with particularly high confidence, but underperformed here, which the authors put down to the absence of established settings for samples this complex.

Taken together, the results show that the choice of acquisition strategy is a primary factor shaping the sensitivity, reproducibility, and functional resolution of a metaproteomic experiment.

3.From Benchmark to Disease Model

To see whether these methodological differences also matter under real biological conditions, the researchers applied the two strongest candidates, DIA- and Slice-PASEF, to a mouse model of inflammatory bowel disease. In these mice, damage to the gut lining heals in the proximal colon but persists in the distal colon, so there is a known biological difference for the methods to find.

The results were remarkably concordant. Both methods identified the same injury-driven shift in the microbial community and the same 182 metabolic pathways in the same 175 bacterial species. On the host side, both detected the same repair signature at the same point in time.

What are the implications of these results for researchers that have to choose between the methods for their studies? The acquisition strategy determines how much a study detects, not what it concludes, so groups can choose the method that fits their logistics, DIA-PASEF for large cohorts, Slice-PASEF when material is scarce, without ending up with a different biological answer. For stool-based diagnostics, where scale is the whole point, that argues for DIA-PASEF.

4.Outlook: Toward Clinical Microbiome Diagnostics

By benchmarking five PASEF strategies within a single, unified framework, this study gives researchers a practical guide for choosing the right method for their specific goals, something the field lacked until now. Combined with the concordant results seen in the mouse model, the work points toward a future where metaproteomics could reliably support both large-scale microbiome research and more targeted clinical applications, such as tracking disease progression or treatment response through stool samples, moving the field closer to real-world, clinically usable microbiome diagnostics.

“Developing better technologies is essential if we want to answer the next generation of questions in microbiome research,” said David Gómez-Varela, who directs the Center of Excellence for Metaproteomics. “We are already applying these advances in international clinical collaborations to study the involvement of microbiomes in neurological, metabolic and autoimmune diseases.”