Total spiking probability edges: A cross-correlation based method for effective connectivity estimation of cortical spiking neurons

Blasi, Stefano De; Ciba, Manuel; Bahmer, Andreas; Thielemann, Christiane

doi:10.1016/j.jneumeth.2018.11.013

Cited by 21 publications

(19 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The finding that functional connections as obtained by CFP follow the rules of spike timing dependent plasticity 16 suggests that thus inferred functional connectivity at least in part reflects synaptic connections between neurons (see Supplementary information for an analysis that supports this connection). Model studies showed that cross-correlation based analyses are far more sensitive to excitatory connections than to inhibitory ones 12 , 17 – 19 , with possible exceptions of sparse networks 20 , or networks with high background activity 21 .…”

Section: Introductionmentioning

confidence: 99%

Maximum entropy models provide functional connectivity estimates in neural networks

Lamberti

Hess

Dias

et al. 2022

Sci Rep

View full text Add to dashboard Cite

Tools to estimate brain connectivity offer the potential to enhance our understanding of brain functioning. The behavior of neuronal networks, including functional connectivity and induced connectivity changes by external stimuli, can be studied using models of cultured neurons. Cultured neurons tend to be active in groups, and pairs of neurons are said to be functionally connected when their firing patterns show significant synchronicity. Methods to infer functional connections are often based on pair-wise cross-correlation between activity patterns of (small groups of) neurons. However, these methods are not very sensitive to detect inhibitory connections, and they were not designed for use during stimulation. Maximum Entropy (MaxEnt) models may provide a conceptually different method to infer functional connectivity. They have the potential benefit to estimate functional connectivity during stimulation, and to infer excitatory as well as inhibitory connections. MaxEnt models do not involve pairwise comparison, but aim to capture probability distributions of sets of neurons that are synchronously active in discrete time bins. We used electrophysiological recordings from in vitro neuronal cultures on micro electrode arrays to investigate the ability of MaxEnt models to infer functional connectivity. Connectivity estimates provided by MaxEnt models correlated well with those obtained by conditional firing probabilities (CFP), an established cross-correlation based method. In addition, stimulus-induced connectivity changes were detected by MaxEnt models, and were of the same magnitude as those detected by CFP. Thus, MaxEnt models provide a potentially powerful new tool to study functional connectivity in neuronal networks.

show abstract

Section: Introductionmentioning

confidence: 99%

Maximum entropy models provide functional connectivity estimates in neural networks

Lamberti

Hess

Dias

et al. 2022

Sci Rep

View full text Add to dashboard Cite

show abstract

“…Any thresholding method should keep only statistically significant connections, i.e., those due to an active connectivity between the considered nodes. In this work, we obtained CMs by means of the Total Spiking Probability Edge (TSPE) algorithm, a correlation-based method, which allows discriminating inhibitory and excitatory connections (De Blasi et al, 2019). Since structural connectivity properties are preserved in the functional CM (Bullmore and Sporns, 2009), a reliable thresholding algorithm should maintain the significant connections of the functional graph, keeping unchanged the topological properties of the structural network as well as the balance of excitatory and inhibitory links.…”

Section: Methodsmentioning

confidence: 99%

“…Taken two spike trains X and Y, we randomly shuffled the timing of each spike in the Y train, keeping constant the total number of spikes (i.e., constant MFR) but varying the interspike temporal interval (i.e., ISI). Once the Y spike train is shuffled, we computed the temporal correlation (De Blasi et al, 2019) between the X-and Y-shuffled spike trains, obtaining a null-case value of connectivity between the neuron X and Y. By iterating such operation a number of time Nshuffling, we were able to obtain a distribution of values that quantify the strength of the functional connectivity between X and Y while the two neurons were not functionally connected.…”

Section: Selected Thresholding Algorithms For Comparison With Ddtmentioning

confidence: 99%

“…One of the principal drawbacks of threshold-based algorithms is that to be conservative (i.e., to not introduce many FPs), a large number of connections is arbitrarily discarded, generating graphs with a small number of connections and altered topological properties. In this work, we tested the DDT method to recover the significant connections of functional CMs obtained by applying a cross-correlation algorithm (De Blasi et al, 2019) to synthetic spike trains generated by the simulation of large-scale neuronal network models. The goodness of the DDT algorithm was evaluated comparing the TCM with the structural one, known a priori since part of the in silico model.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Thresholding Functional Connectivity Matrices to Recover the Topological Properties of Large-Scale Neuronal Networks

2021

View full text Add to dashboard Cite

The identification of the organization principles on the basis of the brain connectivity can be performed in terms of structural (i.e., morphological), functional (i.e., statistical), or effective (i.e., causal) connectivity. If structural connectivity is based on the detection of the morphological (synaptically mediated) links among neurons, functional and effective relationships derive from the recording of the patterns of electrophysiological activity (e.g., spikes, local field potentials). Correlation or information theory-based algorithms are typical routes pursued to find statistical dependencies and to build a functional connectivity matrix. As long as the matrix collects the possible associations among the network nodes, each interaction between the neuron i and j is different from zero, even though there was no morphological, statistical or causal connection between them. Hence, it becomes essential to find and identify only the significant functional connections that are predictive of the structural ones. For this reason, a robust, fast, and automatized procedure should be implemented to discard the “noisy” connections. In this work, we present a Double Threshold (DDT) algorithm based on the definition of two statistical thresholds. The main goal is not to lose weak but significant links, whose arbitrary exclusion could generate functional networks with a too small number of connections and altered topological properties. The algorithm allows overcoming the limits of the simplest threshold-based methods in terms of precision and guaranteeing excellent computational performances compared to shuffling-based approaches. The presented DDT algorithm was compared with other methods proposed in the literature by using a benchmarking procedure based on synthetic data coming from the simulations of large-scale neuronal networks with different structural topologies.

show abstract

“…Units with a spiking rate smaller than 0.3 spikes/s were discarded from the analysis. We used total spiking probability edges (TPSE) algorithm (https://github.com/biomemsLAB/TSPE) (De Blasi et al, 2019) to identify in a computationally efficient manner putative inhibitory connections between units and all clusters recorded. The parameters used were: ▪ d = 0, ▪ neg_wins = [2, 3, 4, 5, 6, 7, 8], ▪ co_wins = 0, ▪ pos_wins = [2, 3, 4, 5, 6], ▪ FLAG_NORM = 1. The connectivity vectors of each unit resulting from TSPE were sorted by inhibition strength.…”

Section: Methodsmentioning

confidence: 99%

Somatostatin interneurons activated by 5-HT_2Areceptor suppress slow oscillations in medial entorhinal cortex

Filippo

Rost

Stumpf

et al. 2020

Preprint

View full text Add to dashboard Cite

1 Serotonin (5-HT) affects multiple physiological processes in the brain and is involved in a 2 number of psychiatric disorders. 5-HT axons reach all cortical areas; however, the precise 3 mechanism by which 5-HT modulates cortical network activity is not yet fully understood. We 4 investigated the effects of 5-HT on slow oscillations (SO), a synchronized cortical network 5 activity universally present across species. SO are observed during slow-wave sleep and 6 anesthesia and are considered the default cortical activity pattern. Combining opto-and 7 pharmacogenetic manipulations with electrophysiological recordings, we discovered that 5-8 HT inhibits SO within the entorhinal cortex (EC) by activating somatostatin-expressing (Som) 9interneurons via the 5-HT2A receptor (5-HT2AR). This receptor is involved in the etiology of 10 different psychiatric disorders and mediates the psychological effects of many psychoactive 11 serotonergic drugs, suggesting that 5-HT targeting of Som interneurons may play an 12 important role in these processes. 13 14 5-HT is one of the most important neuromodulators in the central nervous system. Projections 15 originating from the Raphe nuclei, the brain-stem structure that comprises the majority of 5-16HT releasing neurons in the brain, reach all cortical and sub-cortical area (Descarries et al., 17 2010). Consequentially, it is not surprising that 5-HT is involved in the regulation of a myriad 18 of physiological functions (e.g. circadian rhythm, mood, memory formation, reward encoding, 19 sexual behavior) and psychiatric disorders including depression, autism, schizophrenia, and 20 anxiety disorders (Monti, 2011, Underwood et al., 2018, Hayes and Greenshaw, 2011, Teixeira et al., 21 2018, Uphouse and Guptarak, 2010, Nakai et al., 2017. 22 5-HT levels in the brain are closely linked to the sleep-wake cycle. Activity of serotonergic 23 raphe neurons is increased during wakefulness, decreased during slow-wave sleep and 24 virtually silent during REM sleep (McGinty and Harper, 1976, Oikonomou et al., 2019). Cortical 25 activity is also influenced by the behavioral state of the animal: slow-wave sleep (SWS) is 26 generally associated to "synchronized" patterns of activity, characterized by low-frequency 27 global fluctuations, whereas active wakefulness and REM sleep features "desynchronized" 28 network activity in which low-frequency fluctuations are absent. The shifting of cortical 29 networks between different patterns of activity is controlled, at least in part, by 30 neuromodulators (Lee and Dan, 2012). For instance, Acetylcholine (Ach) can profoundly alter 31 cortical network activity by inducing desynchronization via activation of Som interneurons 32 (Chen et al. , 2015). However, it is not solely responsible for cortical desynchronization as lesions 33 of cholinergic neurons are not sufficient to abolish desynchronization (Kaur et al., 2008). On the 34 other hand, blocking Ach and 5-HT transmission at the same time causes a complete 35 suppression of cortical desynchronizat...

show abstract

Total spiking probability edges: A cross-correlation based method for effective connectivity estimation of cortical spiking neurons

Cited by 21 publications

References 48 publications

Maximum entropy models provide functional connectivity estimates in neural networks

Maximum entropy models provide functional connectivity estimates in neural networks

Thresholding Functional Connectivity Matrices to Recover the Topological Properties of Large-Scale Neuronal Networks

Somatostatin interneurons activated by 5-HT_2Areceptor suppress slow oscillations in medial entorhinal cortex

Contact Info

Product

Resources

About

Total spiking probability edges: A cross-correlation based method for effective connectivity estimation of cortical spiking neurons

Cited by 21 publications

References 48 publications

Maximum entropy models provide functional connectivity estimates in neural networks

Maximum entropy models provide functional connectivity estimates in neural networks

Thresholding Functional Connectivity Matrices to Recover the Topological Properties of Large-Scale Neuronal Networks

Somatostatin interneurons activated by 5-HT2Areceptor suppress slow oscillations in medial entorhinal cortex

Contact Info

Product

Resources

About

Somatostatin interneurons activated by 5-HT_2Areceptor suppress slow oscillations in medial entorhinal cortex