Identifying Neural Signatures of Dopamine Signaling with Machine Learning

Abstract

The emergence of new tools to image neurotransmitters, neuromodulators, and neuropeptides has transformed our understanding of the role of neurochemistry in brain development and cognition, yet analysis of this new dimension of neurobiological information remains challenging. Here, we image dopamine modulation in striatal brain tissue slices with near-infrared catecholamine nanosensors (nIRCat) and implement machine learning to determine which features of dopamine modulation are unique to changes in stimulation strength, and to different neuroanatomical regions. We trained a support vector machine and a random forest classifier to decide whether the recordings were made from the dorsolateral striatum (DLS) versus the dorsomedial striatum (DMS) and find that machine learning is able to accurately distinguish dopamine release that occurs in DLS from that occurring in DMS in a manner unachievable with canonical statistical analysis. Furthermore, our analysis determines that dopamine modulatory signals including the number of unique dopamine release sites and peak dopamine released per stimulation event are most predictive of neuroanatomy. This is in light of integrated neuromodulator amount being the conventional metric used to monitor neuromodulation in animal studies. Lastly, our study finds that machine learning discrimination of different stimulation strengths or neuroanatomical regions is only possible in adult animals, suggesting a high degree of variability in dopamine modulatory kinetics during animal development. Our study highlights that machine learning could become a broadly utilized tool to differentiate between neuroanatomical regions or between neurotypical and disease states, with features not detectable by conventional statistical analysis.

Surface-Enhanced Raman Scattering Nanosensing and Imaging in Neuroscience

“…In such a case, SVMs provide an alternative route to effectively model both linear and nonlinear relationships within the SERS data by finding optimal hyperplanes (decision boundaries) in high-dimensional spectral data spaces. Taking an example from the fluorescence imaging field, this has been particularly useful in distinguishing DA release patterns in neuroanatomical regions like the dorsolateral striatum (DLS) and dorsomedial striatum (DMS), where traditional methods such as LDA fail to capture subtle spectral differences. SVM-driven SERS has been also extended to the identification of Aβ oligomers underlying AD development and stratification of AD patients with ∼85% prediction accuracy .…”

“…Alternatively, Random forests (RFs) can be used to improve classification accuracy by aggregating multiple decision trees, effectively handling high-dimensional data, and modeling complex chemical interactions . This makes RFs suitable for analyzing neurochemical SERS spectral data from proof-of-concept studies where the sample size is usually small, , classifying AD disease stages, and various brain cancers like glioblastomas and gliomas. , …”

Bovine Serum Albumin Template-Mediated Fabrication of Ruthenium Dioxide/Multiwalled Carbon Nanotubes: High-Performance Electrochemical Dopamine Biosensing in Human Serum

“…It not only regulates numerous physiological activities but also serves as an indicator for various disease diagnosis . Abnormal levels of DA may lead to various neurological disorders, such as epilepsy, Parkinson’s disease, Alzheimer’s disease, mania, and schizophrenia. − Moreover, DA as an intravenous medication may result in an elevation of blood pressure and heart rate . Therefore, it is necessary to establish a rapid and accurate sensing platform for detecting DA in clinical and medical diagnostics.…”