Mutual generation in neuronal activity across the brain via deep neural approach, and its network interpretation
Abstract In the brain, many regions work in a network-like association, yet it is not known how durable these associations are in terms of activity and could survive without structural connections. To assess the association or similarity between brain regions with a generating approach, this study e...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2023-10-01
|
| Series: | Communications Biology |
| Online Access: | https://doi.org/10.1038/s42003-023-05453-2 |
| _version_ | 1797636753868193792 |
|---|---|
| author | Ryota Nakajima Arata Shirakami Hayato Tsumura Kouki Matsuda Eita Nakamura Masanori Shimono |
| author_facet | Ryota Nakajima Arata Shirakami Hayato Tsumura Kouki Matsuda Eita Nakamura Masanori Shimono |
| author_sort | Ryota Nakajima |
| collection | DOAJ |
| description | Abstract In the brain, many regions work in a network-like association, yet it is not known how durable these associations are in terms of activity and could survive without structural connections. To assess the association or similarity between brain regions with a generating approach, this study evaluated the similarity of activities of neurons within each region after disconnecting between regions. The “generation” approach here refers to using a multi-layer LSTM (Long Short-Term Memory) model to learn the rules of activity generation in one region and then apply that knowledge to generate activity in other regions. Surprisingly, the results revealed that activity generation from one region to disconnected regions was possible with similar accuracy to generation between the same regions in many cases. Notably, firing rates and synchronization of firing between neuron pairs, often used as neuronal representations, could be reproduced with precision. Additionally, accuracies were associated with the relative angle between brain regions and the strength of the structural connections that initially connected them. This outcome enables us to look into trends governing non-uniformity of the cortex based on the potential to generate informative data and reduces the need for animal experiments. |
| first_indexed | 2024-03-11T12:38:41Z |
| format | Article |
| id | doaj.art-6a4cb90b3ad94aafb90510e63501f030 |
| institution | Directory Open Access Journal |
| issn | 2399-3642 |
| language | English |
| last_indexed | 2024-03-11T12:38:41Z |
| publishDate | 2023-10-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Communications Biology |
| spelling | doaj.art-6a4cb90b3ad94aafb90510e63501f0302023-11-05T12:26:56ZengNature PortfolioCommunications Biology2399-36422023-10-016111410.1038/s42003-023-05453-2Mutual generation in neuronal activity across the brain via deep neural approach, and its network interpretationRyota Nakajima0Arata Shirakami1Hayato Tsumura2Kouki Matsuda3Eita Nakamura4Masanori Shimono5Kyoto University, Graduate School of MedicineKyoto University, Graduate School of MedicineKyoto University, Graduate School of MedicineKyoto University, Graduate School of MedicineKyoto University, Graduate School of InformaticsKyoto University, Graduate School of MedicineAbstract In the brain, many regions work in a network-like association, yet it is not known how durable these associations are in terms of activity and could survive without structural connections. To assess the association or similarity between brain regions with a generating approach, this study evaluated the similarity of activities of neurons within each region after disconnecting between regions. The “generation” approach here refers to using a multi-layer LSTM (Long Short-Term Memory) model to learn the rules of activity generation in one region and then apply that knowledge to generate activity in other regions. Surprisingly, the results revealed that activity generation from one region to disconnected regions was possible with similar accuracy to generation between the same regions in many cases. Notably, firing rates and synchronization of firing between neuron pairs, often used as neuronal representations, could be reproduced with precision. Additionally, accuracies were associated with the relative angle between brain regions and the strength of the structural connections that initially connected them. This outcome enables us to look into trends governing non-uniformity of the cortex based on the potential to generate informative data and reduces the need for animal experiments.https://doi.org/10.1038/s42003-023-05453-2 |
| spellingShingle | Ryota Nakajima Arata Shirakami Hayato Tsumura Kouki Matsuda Eita Nakamura Masanori Shimono Mutual generation in neuronal activity across the brain via deep neural approach, and its network interpretation Communications Biology |
| title | Mutual generation in neuronal activity across the brain via deep neural approach, and its network interpretation |
| title_full | Mutual generation in neuronal activity across the brain via deep neural approach, and its network interpretation |
| title_fullStr | Mutual generation in neuronal activity across the brain via deep neural approach, and its network interpretation |
| title_full_unstemmed | Mutual generation in neuronal activity across the brain via deep neural approach, and its network interpretation |
| title_short | Mutual generation in neuronal activity across the brain via deep neural approach, and its network interpretation |
| title_sort | mutual generation in neuronal activity across the brain via deep neural approach and its network interpretation |
| url | https://doi.org/10.1038/s42003-023-05453-2 |
| work_keys_str_mv | AT ryotanakajima mutualgenerationinneuronalactivityacrossthebrainviadeepneuralapproachanditsnetworkinterpretation AT aratashirakami mutualgenerationinneuronalactivityacrossthebrainviadeepneuralapproachanditsnetworkinterpretation AT hayatotsumura mutualgenerationinneuronalactivityacrossthebrainviadeepneuralapproachanditsnetworkinterpretation AT koukimatsuda mutualgenerationinneuronalactivityacrossthebrainviadeepneuralapproachanditsnetworkinterpretation AT eitanakamura mutualgenerationinneuronalactivityacrossthebrainviadeepneuralapproachanditsnetworkinterpretation AT masanorishimono mutualgenerationinneuronalactivityacrossthebrainviadeepneuralapproachanditsnetworkinterpretation |