Droplet Cyclic Ion Mobility Mass Spec: Compressing Months of Enzyme Screening Into Just 14 Hours

What if you could compress 179 days of grueling laboratory workflow into just 14 hours?

In this episode of ‪ChromatographyTalk‬, ‪Organomation‬ General Manager David Oliva sits down with analytical chemist Laura Penabad to discuss her groundbreaking new research in biocatalysis and high-throughput screening. Evolving engineered enzymes typically requires checking thousands of variants—a massive bottleneck when traditional Liquid Chromatography-Mass Spectrometry (LC-MS) runs take minutes per sample. Laura reveals how her team built a modular screening platform that shatters this limitation, running samples 128 times faster than standard methods without sacrificing analytical accuracy.

Laura breaks down the "orchestra" of technologies making this possible, including coupling nanoliter droplet microfluidics with Cyclic Ion Mobility-Mass Spectrometry (cIM-MS) to separate complex chemical isomers in milliseconds. She also shares the chaotic but triumphant behind-the-scenes stories of her PhD journey—from presenting a "comically bad" early poster at ASMS to recruiting an undergraduate mechanical engineer to build custom automation software from scratch.

If you are an early-career researcher, a mass spec enthusiast, or curious about the future of synthetic biology and machine learning in chemistry, this episode is packed with invaluable intuition and advice.

Video Transcription

From Months to Hours: Droplet Ion Mobility–MS for Biocatalysis Screening

High-throughput screening is central to directed enzyme evolution, but conventional analytical workflows can create a difficult compromise: extremely fast screens often provide limited structural information, while techniques such as LC-MS offer much richer chemical insight at the cost of throughput. In a recent interview, analytical chemist Laura described how her work with droplet-based cyclic ion mobility mass spectrometry is helping bridge that gap—providing rapid analysis while retaining the ability to distinguish structurally similar products.

Her research grew from an initially modest proof of concept into a screening platform capable of analyzing a 96-well plate in roughly ten minutes. The resulting workflow was approximately 128 times faster than a conventional three-minute LC-MS analysis, opening new possibilities for directed evolution campaigns in which thousands of enzyme variants need to be evaluated.

Directed evolution needs better analytical screening

Biocatalysis takes advantage of enzymes as highly selective catalysts capable of performing reactions under relatively mild conditions. Directed evolution extends those capabilities by deliberately modifying enzyme sequences, expressing large numbers of variants, screening their activity, and using the best-performing mutants as starting points for subsequent rounds of evolution.

The process can be repeated multiple times, progressively steering an enzyme toward improved activity, altered substrate specificity, or even chemistry that does not occur naturally.

But the effectiveness of directed evolution depends heavily on the screening step. The more variants that can be analyzed—and the more useful chemical information that can be collected for each one—the more effectively researchers can explore the enormous sequence space available to an enzyme.

For Laura, this created a clear analytical challenge: develop a screening platform that was substantially faster than conventional chromatographic analysis without sacrificing the structural discrimination needed for chemically similar products.

From an imperfect first experiment to a working platform

The project did not begin with a polished high-throughput workflow.

Laura recalled looking back at her 2023 ASMS poster, where the early system was still working with standards rather than true isomeric samples. Droplet introduction was slow—around six seconds per droplet—and the signals were far from the ideal reproducible, square, flat-topped profiles expected from replicate droplets.

Nevertheless, simply producing a droplet signal represented an important first milestone.

The decisive change came when the project stopped being treated merely as a demonstration that droplet ion mobility–MS could work and instead became a deliberate effort to build a robust screening platform.

Laura approached the problem in modules. First came reliable droplet introduction. Then the ion mobility separation method. Quantification became another independent component. Matrix effects were introduced later, gradually making the system resemble the conditions expected with real screening samples.

Each layer was tested before another was added.

This stepwise strategy became one of the central principles of the project: rather than attempting to optimize the entire system simultaneously, individual components were developed, challenged, and then assembled into an increasingly realistic workflow.

Finding the middle ground between speed and structural information

One of the biggest analytical problems in high-throughput screening is distinguishing products with the same nominal or exact mass.

Constitutional isomers may contain the same atoms and therefore generate the same mass-to-charge ratio, while differing only in how functional groups are arranged within the molecular structure. Very rapid optical or mass-based screens may therefore struggle to distinguish them.

Traditional LC-MS can solve many of these problems through chromatographic separation, but a several-minute analysis rapidly becomes impractical when thousands of enzyme variants must be screened.

Ion mobility offered Laura’s team another option.

Rather than separating compounds on a chromatographic timescale, ion mobility can differentiate ions according to their gas-phase behavior on a much shorter timescale. In the workflow described in the interview, this created a useful compromise: not the kilohertz throughput possible with some simple screening technologies, but dramatically faster structural differentiation than conventional LC-MS.

The goal was therefore not simply to achieve the fastest possible screen, but to find a middle-throughput analytical regime with enough resolving power to answer chemically meaningful questions.

A 96-well plate in about ten minutes

That strategy eventually produced one of the most striking results of the project.

The droplet cyclic ion mobility–MS platform processed a 96-well plate in roughly ten minutes, while maintaining quantitative results comparable with the conventional LC-MS workflow used for comparison. According to the interview, this represented an approximately 128-fold increase in analysis speed relative to a three-minute LC-MS run.

Laura described the successful plate experiment less as a surprise than as confirmation that the modular development strategy had worked.

Before reaching that point, the workflow had already been challenged through a sequence of increasingly demanding experiments: droplet formation, ion mobility separation, calibration, matrix-containing calibration, and cross-testing analytes at different concentrations. By the time the complete plate was analyzed, each individual part of the analytical system had already been tested under progressively more realistic conditions.

What does that speed mean for directed evolution?

The impact becomes even clearer when the workflow is considered at the scale of an entire directed-evolution campaign.

Laura’s manuscript estimated that a five-round campaign involving 5,000 variants would require approximately 179 days using the conventional LC-MS workflow but only about 14 hours using the new platform.

Compressing months of analytical work into less than a day has consequences beyond laboratory productivity.

It allows researchers to screen more enzyme variants and therefore sample a larger portion of sequence space. But the resulting datasets may also become increasingly valuable for computational approaches.

Rather than retaining information only about the best-performing enzyme variant, researchers can preserve data describing variants that performed poorly as well as those that succeeded. When genotype and phenotype information remain linked through the well-plate workflow, these larger datasets can potentially provide richer input for machine-learning models designed to guide subsequent rounds of enzyme evolution.

Faster analysis therefore does more than accelerate an existing experiment—it can change the scale and design of the experiment itself.

Balancing ion mobility resolution and signal

Increasing cyclic ion mobility separation time can improve resolution, but it also introduces an important trade-off: additional passes can reduce signal.

Laura optimized this balance experimentally by continuously infusing mixtures containing both isomers and progressively adjusting the ion mobility conditions. After tuning parameters such as wave height and velocity, increasing the separation to three passes provided baseline resolution at the high analyte concentrations used during method development.

Concentration, however, also mattered.

At higher ion populations, space-charge effects could broaden peaks and reduce the baseline separation. This meant that the optimal number of passes could not be selected independently of the concentration range required by the final assay.

Her broader advice for optimization is therefore application-driven: decide first what the analytical method actually needs to accomplish.

More speed will almost always require sacrificing something—resolution, signal intensity, quantitative precision, or another performance parameter. The key is to identify which compromise the application can tolerate and then optimize the system around that requirement.

Solving droplet carryover with a practical compromise

Droplet-based workflows present their own engineering challenges, including carryover between neighboring samples.

In Laura’s system, analytes showed little tendency to partition into the fluorinated oil separating aqueous droplets within the Teflon tubing. The more troublesome carryover occurred closer to electrospray ionization, where a small amount of sample could remain associated with the interface and appear when the next droplet arrived.

Rather than adding excessive engineering complexity to an already elaborate platform, the researchers adopted a pragmatic solution: a sacrificial wash droplet between analytical samples.

It is an example of a recurring theme throughout the project. The objective was not to construct the theoretically most sophisticated droplet system, but to create a screening workflow that remained sufficiently robust, reproducible, and practical for large numbers of samples.

High-throughput analysis also requires high-throughput data processing

Accelerating the mass spectrometry experiment created another bottleneck: the data.

The instrument produced a continuous intensity-versus-time trace rather than a neatly organized well-plate dataset. With small experiments, Laura could manually identify individual droplets and process the results in spreadsheets. That approach quickly became unrealistic as sample numbers increased.

The solution emerged through collaboration with an undergraduate mentee, Aiden, who helped develop a dedicated software tool for processing the continuous traces.

Laura defined how the software should behave and how the analytical information should be organized, while Aiden developed the computational implementation. The workflow itself was subsequently adjusted to make automated data processing more reliable—for example, by introducing marker columns that allowed the software to recognize the end of each plate row.

The resulting tool can accept the acquired traces together with information about plate dimensions and replicates and automatically return parameters including peak identity, duration, and the corresponding well position. A later version also generated an automatically populated heat map.

The experience illustrates how improving analytical throughput often requires simultaneous development of instrumentation, experimental design, and informatics. Making the measurement faster is useful only if the resulting data can be processed at a comparable speed.

Mentoring became part of the scientific process

The software project also reflects another theme Laura emphasized throughout the interview: the value of mentoring.

Several younger researchers contributed directly to the development of the final platform. Laura described mentees not simply as additional hands in the laboratory but as scientists capable of proposing their own hypotheses and experimental ideas.

Working with undergraduate researchers also influenced the way she organized experiments.

Because their available laboratory time was limited, each task needed to have a clear purpose. That encouraged a more disciplined approach to experimental planning: instead of performing experiments simply because they could be done, the team had to ask which experiment would most efficiently answer the next important question.

In that sense, mentoring became another mechanism for improving the research itself.

The next step: continuous, automated screening

Despite the dramatic increase in analytical speed, the current workflow still includes manual operations such as loading tubing and connecting the sample stream to the mass spectrometer.

The next engineering challenge is therefore automation.

Laura described ongoing work using the Venturi effect to pull samples directly from wells into the droplet system. Building on previous work in the laboratory with 96-well plates, she has already tested this approach with as many as 2,000 samples.

The long-term direction is clear: continuous and robust droplet generation, automated handling of thousands of samples, rapid ion mobility separation, mass spectrometric detection, and automated reconstruction of the resulting data into plate-based outputs.

Combining those elements could move the system closer to the type of large-scale screening required for increasingly ambitious biocatalysis campaigns.

Learning mass spectrometry one problem at a time

Laura also reflected on the broader experience of developing the platform during her PhD.

Ion mobility and mass spectrometry initially appeared intimidating, particularly because they combine concepts from analytical chemistry, physical chemistry, separations, instrumentation, and electrical engineering. What eventually changed was not simply learning more facts, but developing the ability to move repeatedly between theory and experiment.

An unexpected result sends the researcher back to the underlying theory. A better understanding suggests a modified experiment. That experiment creates another observation, gradually building intuition about how the system behaves.

Her advice to early-career scientists is therefore to accept that this process takes time.

Advanced analytical techniques may initially appear overwhelming, but understanding develops incrementally through repeated interaction between theory, experimentation, failure, and refinement. Equally important, she emphasized the role of the supportive ion mobility–mass spectrometry community in making a technically demanding field more approachable for young researchers.

From the first inconsistent droplets to a platform capable of processing thousands of samples, that iterative mindset ultimately shaped the entire project.

This text has been automatically transcribed from a video presentation using AI technology. It may contain inaccuracies and is not guaranteed to be 100% correct.

Concentrating on Chromatography Podcast

Dive into the frontiers of chromatography, mass spectrometry, and sample preparation with host David Oliva. Each episode features candid conversations with leading researchers, industry innovators, and passionate scientists who are shaping the future of analytical chemistry. From decoding PFAS detection challenges to exploring the latest in AI-assisted liquid chromatography, this show uncovers practical workflows, sustainability breakthroughs, and the real-world impact of separation science. Whether you’re a chromatographer, lab professional, or researcher you'll discover inspiring content!

You can find Concentrating on Chromatography Podcast in podcast apps:

• Spotify
• Apple podcast
• podscan

and on YouTube channel