Biology subjects

Prokin, I.

Publications and source records attributed to Prokin, I..

3 recordsLinked to original sources

bioRxiv2021

Neural Activity Classification with Machine Learning Models Trained on Interspike Interval Time-Series Data

Modern well-performing approaches to neural decoding are based on machine learning models such as decision tree ensembles and deep neural networks. The wide range of algorithms that can be utilized to learn from neural spike trains, which are essentially time-series data, results in the need for diverse and challenging benchmarks for neural decoding, similar to the ones in the fields of computer vision and natural language processing. In this work, we propose a spike train classification benchmark, based on open-access neural activity datasets and consisting of several learning tasks such as stimulus type classification, animals behavioral state prediction and neuron type identification. We demonstrate that an approach based on hand-crafted time-series feature engineering establishes a strong baseline performing on par with state-of-the-art deep learning based models for neural decoding. We release the code allowing to reproduce the reported results. Author summaryMachine learning-based neural decoding has been shown to outperform the traditional approaches like Wiener and Kalman filters on certain key tasks [1]. To further the advancement of neural decoding models, such as improvements in deep neural network architectures and better feature engineering for classical ML models, there need to exist common evaluation benchmarks similar to the ones in the fields of computer vision or natural language processing. In this work, we propose a benchmark consisting of several individual neuron spike train classification tasks based on open-access data from a range of animals and brain regions. We demonstrate that it is possible to achieve meaningful results in such a challenging benchmark using the massive time-series feature extraction approach, which is found to perform similarly to state-of-the-art deep learning approaches.

neuroscience↗

bioRxiv2019

BDNF controls bidirectional endocannabinoid-plasticity at corticostriatal synapses

The dorsal striatum exhibits bidirectional corticostriatal synaptic plasticity, NMDAR- and endocannabinoids-(eCB)-mediated, necessary for the encoding of procedural learning. Therefore, characterizing factors controlling corticostriatal plasticity is of crucial importance. Brain-derived neurotrophic factor (BDNF) and its receptor, the tropomyosine receptor kinase-B (TrkB), shape striatal functions and their dysfunction deeply affect basal ganglia. BDNF/TrkB signaling controls NMDAR-plasticity in various brain structures including striatum. However, despite cross-talks between BDNF and eCBs, the role of BDNF in eCB-plasticity remains unknown. Here, we show that BDNF/TrkB signaling promotes eCB-plasticity (LTD and LTP) induced by rate-based (low-frequency stimulation) or spike-timing-based (spike-timing-dependent plasticity, STDP) paradigm in striatum. We show that TrkB activation is required for the expression and the scaling of both eCB-LTD and eCB-LTP. Using two-photon imaging of the dendritic spines combined with patch-clamp recordings, we show that TrkB activation induces an intracellular calcium boost, thus increasing eCB synthesis and release. We provide a mathematical model for the dynamics of the signaling pathways involved in corticostriatal plasticity. Finally, we show that TrkB activation allows an enlargement of the domain of expression of eCB-STDP. Our results reveal a novel role for BDNF/TrkB signaling in governing eCB-plasticity expression in striatum, and thus the engram of procedural learning.

neuroscience↗

bioRxiv2018

Robustness of STDP to spike timing jitter

In Hebbian plasticity, neural circuits adjust their synaptic weights depending on patterned firing of action potential on either side of the synapse. Spike-timing-dependent plasticity (STDP) is an experimental implementation of Hebbs postulate that relies on the precise order and the millisecond timing of the paired activities in pre- and postsynaptic neurons. In recent years, STDP has attracted considerable attention in computational and experimental neurosciences. However, canonical STDP is assessed with deterministic (constant) spike timings and time intervals between successive pairings, thus exhibiting a regularity that strongly differs from the biological variability. Hence, the emergence of STDP from noisy neural activity patterns as expected in in vivo-like firing remains unresolved. Here, we used noisy STDP stimulations where the spike timing and/or the interval between successive pairings were jittered. We explored with a combination of experimental neurophysiology and mathematical modeling, the impact of jittering on three distinct forms of STDP at corticostriatal synapses: NMDAR-mediated tLTP, endocannabinoid-mediated tLTD and endocannabinoid-mediated tLTP. As the main result, we found a differential sensitivity to jittered spike timing: NMDAR-tLTP was highly fragile whereas endocannabinoid-plasticity (tLTD and tLTP) appeared more resistant. Moreover, when the frequency or the number of pairings was increased, NMDAR-tLTP became more robust and could be expressed despite strong jittering of the spike timing. Taken together, our results identify endocannabinoid-mediated plasticity as a robust form of STDP while the sensitivity to jitter of NMDAR-tLTP varies with activity frequency. This provides new insights into the mechanisms at play during the different phases of learning and memory and the emergence of Hebbian plasticity in in vivo-like firing.

neuroscience↗