The technique aims to untangle overlapping electrochemical signals to monitor multiple brain chemicals simultaneously during neurochemical monitoring.
Measuring Neurotransmitters with Voltammetry and Machine Learning
Monitoring brain chemistry in real time remains a core challenge in neurochemistry. According to Andrews, traditional methods like microdialysis involve implanting a semi-permeable membrane into the brain’s extracellular space to sample fluid, which is then analyzed offline using high-performance liquid chromatography coupled with electrochemistry. While microdialysis permits multiplexing through chemical separation, it offers limited temporal resolution—reaching roughly one minute per sample in the Andrews lab.
Conversely, voltammetry relies on directly implanting small sensors, commonly carbon fiber microelectrodes, into the tissue. This approach yields sub-second temporal resolution and superior spatial resolution. However, voltammetry historically struggles because electroactive neurotransmitters produce similar, overlapping voltammograms. This overlap complicates efforts to detect multiple substances simultaneously, particularly at low concentrations.
To overcome these limitations, the UCLA group applies machine learning on both ends of the voltammetry pipeline, according to Andrews. On the front end, researchers use Bayesian optimization to rapidly explore large waveform spaces and design custom waveforms tailored for specific detection goals. On the back end, machine learning models analyze large datasets while preserving background current data, allowing algorithms to isolate overlapping electrochemical signals.
Expanding Access with Open-Source Software
To support broader adoption across the research community, the Andrews lab has made its analytical software publicly available. According to Andrews, three of the four modules are currently accessible. Users can run the tools through MATLAB or utilize a standalone version that does not require a MATLAB license.
The complete code has been published on GitHub for advanced users looking to adapt the tools for custom applications. Additionally, the lab published detailed written tutorials and video guides produced by undergraduate students, demonstrating the accessibility of the platform for new users. The software enables researchers to design waveforms in real time and feed hours of recorded data directly into preprocessing modules.
Beyond Dopamine and Serotonin
While much of the lab’s current work centers on dopamine—linked to substance use and movement disorders—and serotonin—implicated in depression and anxiety—the methodology extends to other neurochemicals. According to Andrews, researchers are actively applying voltammetry to analyze other electroactive monoamines, including norepinephrine, epinephrine, and histamine.
Recent studies and presentations also highlight the adaptation of voltammetry for measuring peptides such as oxytocin, endorphins, and met-enkephalin. Expanding the range of detectable transmitters brings the field closer to multiplexed monitoring of complex chemical fluxes in the brain.
Translating Neurochemical Monitoring to Real-World Systems
Moving analytical tools out of controlled laboratory environments and into systems that approximate real-world conditions represents the next major milestone for the field, according to Andrews. Achieving this translation will require closer collaboration between neurochemists, neuroscientists, and clinical researchers who may lack specialized technical instrumentation.
Conferences such as Pittcon provide a vital venue for fostering these interdisciplinary partnerships. According to Andrews, in-person meetings allow researchers to discuss data, establish new collaborations, and evaluate emerging laboratory technologies directly with instrument vendors.
Did You Know? Pittcon serves as North America’s largest annual conference and exposition in laboratory science, with over 90% of its net proceeds directed toward supporting global science education, scholarships, and STEM outreach initiatives.
Frequently Asked Questions
What is the main challenge of using voltammetry for neurotransmitters?
Voltammetry sensors generate overlapping signals because different electroactive neurotransmitters produce similar voltammograms, making it difficult to measure multiple substances at once without advanced pattern-recognition models.
How does Bayesian optimization assist in waveform design?
According to Andrews, researchers use Bayesian optimization to quickly explore massive waveform spaces and converge on entirely new electrochemical waveforms designed for specific detection parameters.
Is the UCLA neurochemical analysis software free to use?
Yes. The software is available via GitHub and MATLAB, and a standalone version is accessible for users who do not own a MATLAB license, accompanied by student-produced video tutorials.
Join the Conversation: What electrochemical sensing challenges is your lab currently tackling? Share your thoughts or questions in the comments below, and explore our archive for more coverage on analytical chemistry innovations.
Worth a look