Conference
Fast reconstruction of degenerate populations of conductance-based neuron models from spike times using deep learning
| Τίτλος: | Fast reconstruction of degenerate populations of conductance-based neuron models from spike times using deep learning |
|---|---|
| Συγγραφείς: | Brandoit, Julien, Ernst, Damien, Drion, Guillaume, Fyon, Arthur |
| Συνεισφορές: | Montefiore Institute - Montefiore Institute of Electrical Engineering and Computer Science - ULiège |
| Πηγή: | NeurIPS 2025 - Workshop : Data on the Brain & Mind, San Diego, United States [US], from 02 December to 07 December 2025. |
| Έτος έκδοσης: | 2025 |
| Θεματικοί όροι: | Quantitative Biology - Neurons and Cognition, Computer Science - Learning, Mathematics - Dynamical Systems, Statistics - Machine Learning, Engineering, computing & technology, Human health sciences, Neurology, Ingénierie, informatique & technologie, Sciences de la santé humaine, Neurologie |
| Περιγραφή: | Inferring the biophysical parameters of conductance-based models (CBMs) from experimentally accessible recordings remains a central challenge in computational neuroscience. Spike times are the most widely available data, yet they reveal little about which combinations of ionic conductances generate the observed activity. This inverse problem is further complicated by neuronal degeneracy, where multiple distinct sets of conductances yield similar spiking patterns. We introduce a method that addresses this challenge by combining deep learning with Dynamic Input Conductances (DICs), a theoretical framework that reduces complex CBMs to three interpretable aggregated conductances that separate according to timescales. DIC values directly relate to excitability and firing patterns. Our approach first maps spike times directly to DIC values at threshold using a lightweight neural network that learns a low-dimensional representation of neuronal activity. The predicted DIC values are then used to generate degenerate CBM populations via an improved state-of-the-art algorithm. Applied to two neuronal models, this algorithmic pipeline reconstructs spiking, bursting, and irregular regimes with high accuracy and robustness to variability, including spike trains generated by Poisson processes. It produces diverse degenerate populations within milliseconds on standard hardware, enabling scalable and efficient inference from spike recordings alone. Beyond methodological advances, we provide an open-source software package with a graphical interface that allows experimentalists to generate and explore CBM populations directly from spike trains without requiring programming expertise. |
| Τύπος εγγράφου: | conference paper http://purl.org/coar/resource_type/c_5794 conferenceObject peer reviewed |
| Γλώσσα: | English |
| Relation: | https://data-brain-mind.github.io/ |
| Σύνδεσμος πρόσβασης: | https://orbi.uliege.be/handle/2268/336182 |
| Rights: | open access http://purl.org/coar/access_right/c_abf2 info:eu-repo/semantics/openAccess |
| Αριθμός Καταχώρησης: | edsorb.336182 |
| Βάση Δεδομένων: | ORBi |
| FullText | Text: Availability: 0 CustomLinks: – Url: https://orbi.uliege.be/handle/2268/336182# Name: EDS - ORBi (ns324271) Category: fullText Text: View record at ORBi |
|---|---|
| Header | DbId: edsorb DbLabel: ORBi An: edsorb.336182 RelevancyScore: 1068 AccessLevel: 3 PubType: Conference PubTypeId: conference PreciseRelevancyScore: 1067.8095703125 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: Fast reconstruction of degenerate populations of conductance-based neuron models from spike times using deep learning – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Brandoit%2C+Julien%22">Brandoit, Julien</searchLink><br /><searchLink fieldCode="AR" term="%22Ernst%2C+Damien%22">Ernst, Damien</searchLink><br /><searchLink fieldCode="AR" term="%22Drion%2C+Guillaume%22">Drion, Guillaume</searchLink><br /><searchLink fieldCode="AR" term="%22Fyon%2C+Arthur%22">Fyon, Arthur</searchLink> – Name: Author Label: Contributors Group: Au Data: Montefiore Institute - Montefiore Institute of Electrical Engineering and Computer Science - ULiège – Name: TitleSource Label: Source Group: Src Data: NeurIPS 2025 - Workshop : Data on the Brain & Mind, San Diego, United States [US], from 02 December to 07 December 2025. – Name: DatePubCY Label: Publication Year Group: Date Data: 2025 – Name: Subject Label: Subject Terms Group: Su Data: <searchLink fieldCode="DE" term="%22Quantitative+Biology+-+Neurons+and+Cognition%22">Quantitative Biology - Neurons and Cognition</searchLink><br /><searchLink fieldCode="DE" term="%22Computer+Science+-+Learning%22">Computer Science - Learning</searchLink><br /><searchLink fieldCode="DE" term="%22Mathematics+-+Dynamical+Systems%22">Mathematics - Dynamical Systems</searchLink><br /><searchLink fieldCode="DE" term="%22Statistics+-+Machine+Learning%22">Statistics - Machine Learning</searchLink><br /><searchLink fieldCode="DE" term="%22Engineering%2C+computing+%26+technology%22">Engineering, computing & technology</searchLink><br /><searchLink fieldCode="DE" term="%22Human+health+sciences%22">Human health sciences</searchLink><br /><searchLink fieldCode="DE" term="%22Neurology%22">Neurology</searchLink><br /><searchLink fieldCode="DE" term="%22Ingénierie%2C+informatique+%26+technologie%22">Ingénierie, informatique & technologie</searchLink><br /><searchLink fieldCode="DE" term="%22Sciences+de+la+santé+humaine%22">Sciences de la santé humaine</searchLink><br /><searchLink fieldCode="DE" term="%22Neurologie%22">Neurologie</searchLink> – Name: Abstract Label: Description Group: Ab Data: Inferring the biophysical parameters of conductance-based models (CBMs) from experimentally accessible recordings remains a central challenge in computational neuroscience. Spike times are the most widely available data, yet they reveal little about which combinations of ionic conductances generate the observed activity. This inverse problem is further complicated by neuronal degeneracy, where multiple distinct sets of conductances yield similar spiking patterns. We introduce a method that addresses this challenge by combining deep learning with Dynamic Input Conductances (DICs), a theoretical framework that reduces complex CBMs to three interpretable aggregated conductances that separate according to timescales. DIC values directly relate to excitability and firing patterns. Our approach first maps spike times directly to DIC values at threshold using a lightweight neural network that learns a low-dimensional representation of neuronal activity. The predicted DIC values are then used to generate degenerate CBM populations via an improved state-of-the-art algorithm. Applied to two neuronal models, this algorithmic pipeline reconstructs spiking, bursting, and irregular regimes with high accuracy and robustness to variability, including spike trains generated by Poisson processes. It produces diverse degenerate populations within milliseconds on standard hardware, enabling scalable and efficient inference from spike recordings alone. Beyond methodological advances, we provide an open-source software package with a graphical interface that allows experimentalists to generate and explore CBM populations directly from spike trains without requiring programming expertise. – Name: TypeDocument Label: Document Type Group: TypDoc Data: conference paper<br />http://purl.org/coar/resource_type/c_5794<br />conferenceObject<br />peer reviewed – Name: Language Label: Language Group: Lang Data: English – Name: NoteTitleSource Label: Relation Group: SrcInfo Data: https://data-brain-mind.github.io/ – Name: URL Label: Access URL Group: URL Data: <link linkTarget="URL" linkTerm="https://orbi.uliege.be/handle/2268/336182" linkWindow="_blank">https://orbi.uliege.be/handle/2268/336182</link> – Name: Copyright Label: Rights Group: Cpyrght Data: open access<br />http://purl.org/coar/access_right/c_abf2<br />info:eu-repo/semantics/openAccess – Name: AN Label: Accession Number Group: ID Data: edsorb.336182 |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsorb&AN=edsorb.336182 |
| RecordInfo | BibRecord: BibEntity: Languages: – Text: English Subjects: – SubjectFull: Quantitative Biology - Neurons and Cognition Type: general – SubjectFull: Computer Science - Learning Type: general – SubjectFull: Mathematics - Dynamical Systems Type: general – SubjectFull: Statistics - Machine Learning Type: general – SubjectFull: Engineering, computing & technology Type: general – SubjectFull: Human health sciences Type: general – SubjectFull: Neurology Type: general – SubjectFull: Ingénierie, informatique & technologie Type: general – SubjectFull: Sciences de la santé humaine Type: general – SubjectFull: Neurologie Type: general Titles: – TitleFull: Fast reconstruction of degenerate populations of conductance-based neuron models from spike times using deep learning Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Brandoit, Julien – PersonEntity: Name: NameFull: Ernst, Damien – PersonEntity: Name: NameFull: Drion, Guillaume – PersonEntity: Name: NameFull: Fyon, Arthur – PersonEntity: Name: NameFull: Montefiore Institute - Montefiore Institute of Electrical Engineering and Computer Science - ULiège IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 01 Type: published Y: 2025 |
| ResultId | 1 |