Biology subjects

Middya, S.

Publications and source records attributed to Middya, S..

3 recordsLinked to original sources

bioRxiv2026

Neural correlates of glioma progression using implanted neural interfaces

High-grade glioma is an incurable brain cancer with a median survival of approximately 14 months. Over the last 50 years, small improvements in patient outcomes have been overshadowed by significant progress in most other cancers. Yet, emerging research has revealed that neural circuits play an active and central role in driving glioma growth and proliferation, highlighting the nervous system as a promising avenue for both disease monitoring and therapeutic intervention. Here, we present a platform for chronically monitoring tumor progression using neural recordings in freely behaving mice with gliomas. Using this platform across multiple mouse strains and glioma models, we show that neural recordings can accurately track tumor progression in vivo. Cancer progression was consistently associated with elevated gamma-band neural activity in the tumor microenvironment across both adult glioblastoma (GBM) and pediatric diffuse intrinsic pontine glioma (DIPG) cancer models. Interestingly, lower frequency neural activity exhibited distinct, cell-line specific changes over time: GBM models exhibited decreases in low frequency neural activity whereas DIPG models exhibited increases over time. Finally, using machine learning models applied to chronic neural recordings from tumor-bearing mice treated with or without standard-of-care chemotherapy (temozolomide for GBM), we accurately predicted tumor burden as inferred through in vivo bioluminescence imaging. By fitting low-dimensional mathematical models to gamma-band neural trajectories, we could further predict individual tumor growth rates over a 5-week period with high accuracy. These results establish that pathological neural-tumor interactions can be harnessed to monitor glioma progression in vivo. Coupling this monitoring capability with therapeutic electrical stimulation in the same device could open up a new class of implantable, closed-loop neurotechnologies with the potential to transform glioma treatment.

neuroscience↗

bioRxiv2025

NeuroSuite for Long-term Functional and Structural Studies of Air-Liquid Interface Cerebral Organoids

Over the past decade, air-liquid interface cerebral organoids (ALI-COs) have emerged as powerful in vitro models that capture essential structural and functional traits of the human brain, offering an exciting alternative to traditional animal models in neuroscience. Yet, the full potential of these systems has remained untapped due to the lack of non-invasive, long-term electrophysiological tools capable of preserving organoid integrity. Existing techniques, ranging from patch clamping to rigid and 3D microelectrode arrays, often compromise organoid growth and disrupt delicate cytoarchitecture. Here, we present NeuroSuite, an innovative bioelectronic platform designed to overcome these challenges. At its core is Neuroweb, a perforated, ultra-thin, and conformable organic microelectrode array engineered for minimal disruption of nutrient and oxygen exchange. Neuroweb is reusable and supports stable recordings for over six months, making it uniquely suited for longitudinal studies. Coated with poly(3,4-ethylenedioxythiophene):polystyrene sulfonate (PEDOT:PSS), a high-performance mixed ionic-electronic conductor, Neuroweb delivers exceptional signal-to-noise ratio recordings with high spatial precision. By pairing Neuroweb with NeuroMaps, an intuitive software for interactive analysis and visualisation, NeuroSuite enables long-term, non-invasive tracking and spatial mapping of electrical activity from brain organoids and ex vivo brain slices at the air-liquid interface. Following rigorous validation, we demonstrate that NeuroSuite can capture both high- and low-frequency throughout maturation. Our pipeline reveals evolving network connectivity, including the development of GABA-ergic interneurons, and concurrent shifts in high-frequency spiking and low-frequency oscillations indicative of a refinement in the excitatory-inhibitory balance. Finally, automated data acquisition and spatial spike mapping highlight local activity changes in response to media composition, a factor often overlooked in conventional recordings. NeuroSuite thus opens a new frontier in organoid neuroscience, enabling precise, long-term monitoring essential for modelling neurological diseases, understanding human brain development, and accelerating drug discovery. TeaserConformal organic bioelectronic arrays, combined with an open-access toolbox for analysis and visualisation, reveal real-time and long-term neural dynamics in brain organoid slices at the air-liquid interface.

neuroscience↗

bioRxiv2024

Machine learning-based spike sorting reveals how subneuronal concentrations of monomeric Tau cause a loss in excitatory postsynaptic currents in hippocampal neurons

Extracellular recordings of neuronal activity constitute a powerful tool for investigating the intricate dynamics of neural networks and the activity of individual neurons. Microelectrode arrays (MEAs) allow for recordings with a high electrode count, ranging from 10s to 1000s, generating extensive datasets of neuronal information. Furthermore, MEAs capture extracellular field potentials from cultured cells, resulting in highly complex neuronal signals that necessitate precise spike sorting for meaningful data extraction. Nevertheless, conventional spike sorting methods face limitations in recognising diverse spike shapes, thereby constraining the full utilisation of the rich dataset acquired from MEA recordings. To overcome these limitations, we have developed a machine learning algorithm, named PseudoSort, which employs advanced self-supervised learning techniques, a distinctive density-based pseudo-labelling strategy, and an iterative fine-tuning process to enhance spike sorting accuracy. Through extensive benchmarking on large-scale simulated datasets, we demonstrate the superior performance of PseudoSort compared to recently developed machine learning-based (ML) spike sorting algorithms. We showcase the practical application of PseudoSort by utilising MEA recordings from hippocampal neurons exposed to subneuronal concentrations of monomeric Tau, a protein associated with Alzheimers disease (AD). Our results, validated against patch clamp experiments, unveil that monomeric Tau at subneuronal concentrations induces stimulation-dependent disruptions in both local and global activity of hippocampal neurons. Remarkably, patch clamp electrophysiology highlights the effect of combined Tau and neuronal stimulation treatment on excitatory postsynaptic currents, whereas PseudoSort excels in identifying neuronal clusters that exhibit diminished firing capacity following Tau treatment alone, i.e., in the absence of stimulation. This comprehensive approach validates the prowess of PseudoSort and unravels the intricate effects of Tau on neuronal activity, particularly in the context of AD.

neuroscience↗