Abstract
dc:descriptionFor a long time, neuroscientists have been on a quest to understand the functional role of individual neurons. This requires understanding the neuron both as a biophysical system governed by its morphology and ion channels, and as a functional unit within the behavior of the whole organism. Dendrites sit at the heart of this endeavor, since it is there that incoming signals are actively reshaped before reaching the soma to trigger the output spikes. Detailed biophysical models with active dendrites have been instrumental in exploring this integration. However, such models can be challenging to build, validate, and interpret due to the large number of parameters involved. To address this, we introduce DendroTweaks, a toolbox designed to make detailed biophysical models with active dendrites more intuitive and more interactive. DendroTweaks features a web-based graphical interface, where users can explore single-cell neuronal models and adjust their morphological and biophysical parameters with real-time visual feedback. The toolbox focuses on subcellular properties and supports a range of experimental protocols that illuminate how these properties shape neuronal activity. It helps users build high-level, modular model representations and includes a rich set of tools for parsing, generating, and standardizing commonly used neuronal data formats. By combining extensive visualization capabilities and comprehensive data management functionality, DendroTweaks introduces a novel interactive approach for unraveling dendritic dynamics. Understanding neuronal function, however, also requires placing the neuron in a relevant context. An evolutionary perspective suggests that this context is a closed-loop interaction with an environment. We therefore propose a framework in which a network of simplified biophysical neurons learns to solve a task through reinforcement learning. Within this framework the inputs to every neuron carry genuine functional task-related content. Taken together, the two contributions open a path to asking how the subcellular machinery of a neuron contributes to a behavioral outcome, which is ultimately the question neuroscientists want to understand.
Degree
thesis:*- Grantor dc:publisher
- University of Crete (UOC)
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Makarov, Roman
- Μακάροφ, Ρομάν
Subjects
dc:subject × 16Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/62742