Biology subjects

Andrade-Ortega, B.

Publications and source records attributed to Andrade-Ortega, B..

2 recordsLinked to original sources

bioRxiv2025

Relevance of Nonlinear Dimensionality Reduction for Efficient and Robust Spike Sorting

Spike sorting is one of the cornerstones of extracellular electrophysiology. By leveraging advanced signal processing and data analysis techniques, spike sorting makes it possible to detect, isolate, and map single neuron spiking activity from both in vivo and in vitro extracellular electrophysiological recordings. A crucial step of any spike sorting pipeline is to reduce the dimensionality of the recorded spike waveform data. Reducing the dimensionality of the processed data is a near-universal practice, fundamentally motivated by the use of clustering algorithms responsible to detect, isolate, and sort the recorded putative neurons. In this paper we propose and illustrate on both synthetic and experimental data that employing the nonlinear dimensionality reduction technique Uniform Manifold Approximation and Projection (UMAP) can drastically improve the performance, efficiency, robustness, and scalability of spike sorting pipelines without increasing their computational cost. We show how replacing the linear or ad hoc, expert-defined, supervised nonlinear dimensionality reduction methods commonly used in spike sorting pipelines by the unsupervised, mathematically grounded, nonlinear dimensionality reduction method provided by UMAP drastically increases the number of correctly sorted neurons, makes the identification of quieter, seldom spiking neurons more reliable, enables deeper and more precise explorations and analysis of the neural code, and paves new ways toward more efficient and end-to-end automatable spike sorting pipelines of large-scale extracellular neural recording as those produced by high-density multielectrode arrays.

neuroscience↗

bioRxiv2024

Multi-Stable Bimodal Perceptual Coding within the Ventral Premotor Cortex

Neurons in the primate ventral premotor cortex (VPC) respond to both tactile and acoustic stimuli, yet how they integrate and process information from these sensory modalities remains unclear. To investigate this, we recorded VPC neuronal activity in two trained monkeys performing a bimodal detection task (BDT), in which they reported the presence or absence of either a tactile or an acoustic stimulus. Single-cell analyses revealed diverse response types, including purely tactile, purely acoustic, bimodal, and neurons with sustained activity during the decision maintenance delay--the period between stimulus offset and motor response. To further examine VPCs role in the BDT, we applied dimensionality reduction techniques to uncover low-dimensional latent dynamics in the neuronal population and conducted parallel analyses using a recurrent neural network (RNN) model trained on the same task. Neural trajectories for tactile and acoustic responses diverged sharply, whereas in stimulus-absent trials, the dynamics remained at rest. During the delay period, the trajectories exhibited a pronounced rotational dynamic, shifting toward a subspace orthogonal to the sensory response space, suggesting memory maintenance through stable equilibria. This indicates that network dynamics can sustain distinct stable states corresponding to the three potential task outcomes. Using low-dimensional modeling, we propose a universal dynamical mechanism that underlies the transition from sensory processing to memory retention, aligning with both experimental and computational findings. These results demonstrate that VPC neurons encode bimodal information, integrate competing sensory inputs, and maintain decisions across the delay period, regardless of sensory modality.

neuroscience↗