News Medical speaks with Anne Andrews of UCLA about measuring neurotransmitters in the brain using voltammetry and machine learning. Andrews explains how her group uses Bayesian optimization to design new voltammetry waveforms and pattern-recognition models to separate overlapping electrochemical signals, with the goal of measuring multiple neurotransmitters at once. She also discusses her lab’s publicly available software and what it could mean for translating neurochemical monitoring from research settings toward real-world use.

To start, what inspired you to pursue a career in science, and how did you end up working at the intersection of analytical chemistry and neuroscience?

Science really started for me early on, going back to middle school and high school, when I had some really great chemistry teachers. So a shoutout to those folks who make that their life’s work: you do influence us.

I went on to earn an undergraduate degree in chemistry and then worked in industry first. My first job was in analytical chemistry.

I knew late in my undergraduate years, and then in that first job, that I really was a measurer as far as chemistry was concerned. I loved to measure things.

Then, when I started a journey back to school to get a PhD, I took a one-year temporary appointment at the National Institute of Mental Health because I wanted a short immersion in basic research to see if I really liked it before committing to the PhD. That’s where I fell in love with neuroscience.

Since then, I’ve been trying to marry the two.

We’ll focus on serotonin and dopamine today. What do we know about their normal roles in the brain, and the conditions they’re linked to?

I’ll tell you what we know. Obviously, there’s a lot we don’t know, but dopamine and serotonin are implicated in both normal behavior and disordered behavior associated with a number of neuropsychiatric conditions.

Disorders such as depression and anxiety are associated with serotonin, and substance use and movement disorders are associated with dopamine. However, they also work together, and we’re starting to learn more about that.

Before we get to the machine-learning piece, how have researchers measured these neurotransmitters, and how do microdialysis and voltammetry compare?

We started early on, really predating my career and going into my early career, by simply being able to measure their concentrations in brain tissue ex vivo.

Then the field developed methods to start monitoring their concentrations in vivo in real time. There are two widely employed methods: the first is microdialysis and the second is voltammetry, both of which we practice in our lab, and I have delivered talks on.

Microdialysis is a technique that involves implanting a semi-permeable membrane, a small device, into the brain, sampling the brain's extracellular space, and then taking those samples offline and analyzing them by an instrumental method.

In my case, we use high-performance liquid chromatography coupled with electrochemistry. This imparts, because of the separation, our ability to multiplex.

With the voltammetry technique, one advantage is that the sensor can be directly implanted. The measurements are made directly on something, commonly a carbon fiber microelectrode, and because this sensor is much smaller, we have better spatial resolution. The temporal resolution is on the order of sub-second, compared to microdialysis, which, in our case, is among the fastest recorded at a minute.

However, voltammetry suffers from the fact that electroactive neurotransmitters have similar voltammograms that overlap. That makes it difficult to examine more than one neurotransmitter at a time and to detect neurotransmitters present at very low concentrations.

The two methods have their own strengths and weaknesses in terms of timescales, spatial resolution, and the ability to multiplex. In my group, we’ve always done both and tried to take advantage of those differences.

What do you see as the big advantage of measuring neurotransmitters in real time, rather than only taking snapshots?

The key advantage, particularly as we improve the scales over which we can measure, is trying to decode how information resides in the chemical fluxes in the brain’s extracellular space. We’re not there yet, but neuroscientists, and neurochemists in particular, are inching closer and closer.

Where does machine learning come into your workflow, and what has it changed for you?

When I give talks about machine learning, I see them as a call to arms, or a call to action, in the field, because we think we’ve developed some ways of using machine learning on both ends of the voltammetry development pipeline.

On the front end, we use Bayesian optimization to explore very large waveform spaces very quickly and converge on totally new waveforms with the specific characteristics we want to employ.

On the back end, we use machine learning. People now commonly call it AI, but there’s a long history of math that underlies this. That enables us to analyze really large data sets, which we generate with every experiment, and to keep all the data in a voltammogram, including the background current. It also allows the machine learning model to find patterns in the data.

This lets us see multiple things that have overlapping voltammograms at the same time.

You’ve also made software available to other researchers. What can people access today, and how do they use it?

Yes, I’m glad you asked. We make our software publicly available; three of the four modules have already been released.

They’re in MATLAB, and we also have a standalone version for the average user. You don’t need to have a MATLAB license to use the software. We’ve also released all the MATLAB code on GitHub, so the super user can adapt the code to whatever their specific use cases might be.

In the supplemental information for the accompanying paper, we published tutorials, and we also have video tutorials made by our undergraduates. We trained undergraduates on this software, and they picked it up very quickly, so anybody should be able to use it.

It’s really flexible software. You can design any conceivable waveform using it, and it gets imported directly into the analysis module. You can change the waveform in real time when you’re doing your analyses.

All your data, which can now be recorded over hours, is fed directly into module three for the preprocessing analysis. And we’ll be dropping module four soon, which will let you work with a number of different machine learning models. We’re really excited about what this will enable the field to do, and people should watch out for the drop of that final module.

Beyond dopamine and serotonin, where else do you think these methods could be useful?

There are a number of other electrochemically active monoamine neurotransmitters that folks have analyzed using voltammetry, beyond serotonin and dopamine. These include: norepinephrine, epinephrine, and histamine.

We’re also seeing some really nice work coming out on using voltammetry to analyze peptides. We’ve heard talks recently and read papers based on research into oxytocin, endorphins, and met-enkephalin, and others. I think the number of transmitters that will be usable, and hopefully then multiplexed, is growing.

More broadly, what would you like to see next as analytical chemists and neuroscientists collaborate on these problems?

What I would like to see more than anything is for these techniques to be used increasingly in translation. I think that’s a two-way street.

One can envision that neurochemists will tackle these important and complex problems in neuroscience and psychiatry more and more by learning about them, but also through collaboration with folks in those fields who aren’t practitioners and maybe don’t have the technical capabilities to use these kinds of techniques.

Moving things out of the lab, out of the flow cell, and out of model systems into systems that more approximate real-world conditions will be the true test of these methods, and of the power they may have.

We’re here at Pittcon 2026 in San Antonio, and you’ve been coming back year after year. What keeps you returning to Pittcon?

Oh my goodness, so many things! First of all, I get to see so many colleagues here in person. One thing the pandemic really brought home for me (and probably for many others) is that it’s one thing to give a talk on Zoom, when you’re talking to outer space; it’s another thing to give a talk to a room full of people who are engaged and asking you questions.

You’re bumping into your colleagues in the hallways. You’re having those real-time conversations, setting up new collaborations, sharing information in a way that you just don’t do when you meet online. Pittcon is one of the premier meetings where not only analytical chemists but also neurochemists gather. So that’s kept me coming back year after year.

Of course, the trade show is always wonderful. I always have fun seeing new instruments and talking to the vendors themselves. It’s just renewing, you know, one’s career every year to come here for a refresh.

Measuring the Brain in Real Time: Neurotransmitters, Voltammetry & AI with Anne Andrews of UCLA

About Anne Andrews

Anne Andrews is based at UCLA, where her research focuses on measuring neurotransmitters in vivo using electrochemical methods such as voltammetry, alongside computational approaches to analyze complex signals. Her work includes developing tools to support real-time neurochemical monitoring and sharing software resources with the wider research community.

About Pittcon

Pittcon is North America’s largest conference and exposition in laboratory science, attracting thousands of industry, academia, and government attendees and exhibitors from around the world in. Pittcon goes beyond the typical conference; it's a global community dedicated to advancing science through education, collaboration, and philanthropy. By bringing together scientists, researchers, educators, students, and industry leaders, Pittcon creates opportunities to learn, connect, and inspire innovation. Beyond the event itself, more than 90% of Pittcon's net proceeds support science education and outreach initiatives locally and globally, funding scholarships, grants, STEM programs, laboratory improvements, and other efforts that strengthen communities and expand access to science.

Pittcon delivers world-class educational and professional development opportunities through its Technical Program, Short Courses, and Networking Workshops and Roundtables. Attendees can explore the latest scientific research through symposia, oral and poster presentations, enhance their skills with expert-led deep-dive courses, and connect with peers through interactive workshops and networking experiences designed to foster collaboration and career growth.

The Pittcon Expo brings the scientific community together in an immersive environment where attendees can discover the latest laboratory technologies, meet with leading instrument and service providers, and experience science in action. Beyond the exhibit booths, the Expo Floor features engaging attractions, educational experiences, demonstrations, and networking spaces, while receptions, parties, and social events create opportunities to build lasting connections with colleagues from around the world.