Spatial Embedding Imposes Constraints on Neuronal Network Architectures

Trends in Cognitive Sciences, Journal Year: 2018, Volume and Issue: 22(12), P. 1127 - 1142

Published: Oct. 26, 2018

Language: Английский

---

Learning function from structure in neuromorphic networks

Nature Machine Intelligence, Journal Year: 2021, Volume and Issue: 3(9), P. 771 - 786

Published: Aug. 9, 2021

Language: Английский

---

A multi-scale cortical wiring space links cellular architecture and functional dynamics in the human brain

PLoS Biology, Journal Year: 2020, Volume and Issue: 18(11), P. e3000979 - e3000979

Published: Nov. 30, 2020

The vast net of fibres within and underneath the cortex is optimised to support convergence different levels brain organisation. Here, we propose a novel coordinate system human based on an advanced model its connectivity. Our approach inspired by seminal, but so far largely neglected models cortico-cortical wiring established postmortem anatomical studies capitalises cutting-edge in vivo neuroimaging machine learning. new expands currently prevailing diffusion magnetic resonance imaging (MRI) tractography incorporation additional features cortical microstructure proximity. Studying several datasets parcellation schemes, could show that our robustly recapitulates sensory-limbic anterior-posterior dimensions A series validation experiments showed space reflects microcircuit (including pyramidal neuron depth glial expression) allowed for competitive simulations functional connectivity dynamics resting-state (rs-fMRI) intracranial electroencephalography (EEG) coherence. results advance understanding how cell-specific neurobiological gradients produce hierarchical scheme concordant with increasing sophistication evaluations demonstrate bridges across scales neural organisation can be easily translated single individuals.

Language: Английский

---

Characterizing cycle structure in complex networks

Communications Physics, Journal Year: 2021, Volume and Issue: 4(1)

Published: Dec. 20, 2021

Abstract A cycle is the simplest structure that brings redundant paths in network connectivity and feedback effects dynamics. An in-depth understanding of which cycles are important what role they play on dynamics, however, still lacking. In this paper, we define number matrix, a matrix enclosing information about network, ratio, an index quantifies node importance. Experiments real networks suggest ratio contains rich addition to well-known benchmark indices. For example, rankings by largely different from degree, H-index, coreness, very similar Numerical experiments identifying vital nodes for synchronization maximizing early reach spreading show performs overall better than other benchmarks. Finally, highlight significant difference between distribution shorter model networks. We believe our analyses may yield insights, metrics, models, algorithms science.

Language: Английский

---

Topological data analysis of zebrafish patterns

Proceedings of the National Academy of Sciences, Journal Year: 2020, Volume and Issue: 117(10), P. 5113 - 5124

Published: Feb. 25, 2020

Self-organized pattern behavior is ubiquitous throughout nature, from fish schooling to collective cell dynamics during organism development. Qualitatively these patterns display impressive consistency, yet variability inevitably exists within pattern-forming systems on both microscopic and macroscopic scales. Quantifying measuring features can inform the underlying agent interactions allow for predictive analyses. Nevertheless, current methods analyzing that arise only capture features, or rely either manual inspection smoothing algorithms lose agent-based nature of data. Here we introduce based topological data analysis interpretable machine learning quantifying agent-level global attributes a large scale. Because zebrafish model skin formation, focus specifically its as means illustrating our approach. Using recent model, simulate thousands wild-type mutant apply methodology better understand in zebrafish. Our able quantify differential impact stochasticity patterns, use predict stripe spot statistics function varying cellular communication. work provides new approach automatically biological so now answer critical questions formation at much larger

Language: Английский

---