Using deep learning to quantify neuronal activation from single-cell and spatial transcriptomic data

E. Bahl; S. Chatterjee; U. Mukherjee; M. Elsadany; Y. Vanrobaeys; L.-C. Lin; M. McDonough; J. Resch; K.P. Giese; T. Abel; J.J. Michaelson

doi:10.1038/s41467-023-44503-5

Using deep learning to quantify neuronal activation from single-cell and spatial transcriptomic data

E. Bahl, S. Chatterjee, U. Mukherjee, M. Elsadany, Y. Vanrobaeys, L.-C. Lin, M. McDonough, J. Resch, K.P. Giese, T. Abel, J.J. Michaelson

Basic and Clinical Neuroscience

Research output: Contribution to journal › Article › peer-review

Abstract

Neuronal activity-dependent transcription directs molecular processes that regulate synaptic plasticity, brain circuit development, behavioral adaptation, and long-term memory. Single cell RNA-sequencing technologies (scRNAseq) are rapidly developing and allow for the interrogation of activity-dependent transcription at cellular resolution. Here, we present NEUROeSTIMator, a deep learning model that integrates transcriptomic signals to estimate neuronal activation in a way that we demonstrate is associated with Patch-seq electrophysiological features and that is robust against differences in species, cell type, and brain region. We demonstrate this method’s ability to accurately detect neuronal activity in previously published studies of single cell activity-induced gene expression. Further, we applied our model in a spatial transcriptomic study to identify unique patterns of learning-induced activity across different brain regions in male mice. Altogether, our findings establish NEUROeSTIMator as a powerful and broadly applicable tool for measuring neuronal activation, whether as a critical covariate or a primary readout of interest.

Original language	English
Article number	779
Journal	Nature Communications
Volume	15
Issue number	1
DOIs	https://doi.org/10.1038/s41467-023-44503-5
Publication status	Published - 26 Jan 2024

Access to Document

10.1038/s41467-023-44503-5Licence: CC BY

https://www.scopus.com/inward/record.uri?eid=2-s2.0-85183328270&doi=10.1038%2fs41467-023-44503-5&partnerID=40&md5=3304899afccb7023696844174e42cc7b

Cite this

@article{0cf6ab0a773b4a09a9c2c8bde4961f4a,

title = "Using deep learning to quantify neuronal activation from single-cell and spatial transcriptomic data",

abstract = "Neuronal activity-dependent transcription directs molecular processes that regulate synaptic plasticity, brain circuit development, behavioral adaptation, and long-term memory. Single cell RNA-sequencing technologies (scRNAseq) are rapidly developing and allow for the interrogation of activity-dependent transcription at cellular resolution. Here, we present NEUROeSTIMator, a deep learning model that integrates transcriptomic signals to estimate neuronal activation in a way that we demonstrate is associated with Patch-seq electrophysiological features and that is robust against differences in species, cell type, and brain region. We demonstrate this method{\textquoteright}s ability to accurately detect neuronal activity in previously published studies of single cell activity-induced gene expression. Further, we applied our model in a spatial transcriptomic study to identify unique patterns of learning-induced activity across different brain regions in male mice. Altogether, our findings establish NEUROeSTIMator as a powerful and broadly applicable tool for measuring neuronal activation, whether as a critical covariate or a primary readout of interest.",

author = "E. Bahl and S. Chatterjee and U. Mukherjee and M. Elsadany and Y. Vanrobaeys and L.-C. Lin and M. McDonough and J. Resch and K.P. Giese and T. Abel and J.J. Michaelson",

note = "Funding Information: This work was supported by NIH grant R01 MH 087463 to T.A., NIH grant R01 DC014489 to J.J.M., NIH grant K99 AG 068306 and the Nellie Ball Trust to S.C., and The University of Iowa Hawkeye Intellectual and Developmental Disabilities Research Center (HAWK-IDDRC) P50 HD103556 to T.A. and Lane Strathearn. T.A. and J.J.M. are also supported by the Roy J. Carver Charitable Trust. We thank the Allen Institute for Brain Sciences for their valuable datasets we used to train our model. We thank the creators and authors of DCA, whose work inspired the approach we implemented in this paper. We also thank Mahesh Shetty for his valuable contributions to discussions about this project. Figure 1 and Supplementary Fig. 10 were created using graphical elements from BioRender.com. Funding Information: This work was supported by NIH grant R01 MH 087463 to T.A., NIH grant R01 DC014489 to J.J.M., NIH grant K99 AG 068306 and the Nellie Ball Trust to S.C., and The University of Iowa Hawkeye Intellectual and Developmental Disabilities Research Center (HAWK-IDDRC) P50 HD103556 to T.A. and Lane Strathearn. T.A. and J.J.M. are also supported by the Roy J. Carver Charitable Trust. We thank the Allen Institute for Brain Sciences for their valuable datasets we used to train our model. We thank the creators and authors of DCA, whose work inspired the approach we implemented in this paper. We also thank Mahesh Shetty for his valuable contributions to discussions about this project. Figure and Supplementary Fig. were created using graphical elements from BioRender.com. Publisher Copyright: {\textcopyright} 2024, This is a U.S. Government work and not under copyright protection in the US; foreign copyright protection may apply.",

year = "2024",

month = jan,

day = "26",

doi = "10.1038/s41467-023-44503-5",

language = "English",

volume = "15",

journal = "Nature Communications",

issn = "2041-1723",

publisher = "Nature Research",

number = "1",

}

TY - JOUR

T1 - Using deep learning to quantify neuronal activation from single-cell and spatial transcriptomic data

AU - Bahl, E.

AU - Chatterjee, S.

AU - Mukherjee, U.

AU - Elsadany, M.

AU - Vanrobaeys, Y.

AU - Lin, L.-C.

AU - McDonough, M.

AU - Resch, J.

AU - Giese, K.P.

AU - Abel, T.

AU - Michaelson, J.J.

N1 - Funding Information: This work was supported by NIH grant R01 MH 087463 to T.A., NIH grant R01 DC014489 to J.J.M., NIH grant K99 AG 068306 and the Nellie Ball Trust to S.C., and The University of Iowa Hawkeye Intellectual and Developmental Disabilities Research Center (HAWK-IDDRC) P50 HD103556 to T.A. and Lane Strathearn. T.A. and J.J.M. are also supported by the Roy J. Carver Charitable Trust. We thank the Allen Institute for Brain Sciences for their valuable datasets we used to train our model. We thank the creators and authors of DCA, whose work inspired the approach we implemented in this paper. We also thank Mahesh Shetty for his valuable contributions to discussions about this project. Figure 1 and Supplementary Fig. 10 were created using graphical elements from BioRender.com. Funding Information: This work was supported by NIH grant R01 MH 087463 to T.A., NIH grant R01 DC014489 to J.J.M., NIH grant K99 AG 068306 and the Nellie Ball Trust to S.C., and The University of Iowa Hawkeye Intellectual and Developmental Disabilities Research Center (HAWK-IDDRC) P50 HD103556 to T.A. and Lane Strathearn. T.A. and J.J.M. are also supported by the Roy J. Carver Charitable Trust. We thank the Allen Institute for Brain Sciences for their valuable datasets we used to train our model. We thank the creators and authors of DCA, whose work inspired the approach we implemented in this paper. We also thank Mahesh Shetty for his valuable contributions to discussions about this project. Figure and Supplementary Fig. were created using graphical elements from BioRender.com. Publisher Copyright: © 2024, This is a U.S. Government work and not under copyright protection in the US; foreign copyright protection may apply.

PY - 2024/1/26

Y1 - 2024/1/26

N2 - Neuronal activity-dependent transcription directs molecular processes that regulate synaptic plasticity, brain circuit development, behavioral adaptation, and long-term memory. Single cell RNA-sequencing technologies (scRNAseq) are rapidly developing and allow for the interrogation of activity-dependent transcription at cellular resolution. Here, we present NEUROeSTIMator, a deep learning model that integrates transcriptomic signals to estimate neuronal activation in a way that we demonstrate is associated with Patch-seq electrophysiological features and that is robust against differences in species, cell type, and brain region. We demonstrate this method’s ability to accurately detect neuronal activity in previously published studies of single cell activity-induced gene expression. Further, we applied our model in a spatial transcriptomic study to identify unique patterns of learning-induced activity across different brain regions in male mice. Altogether, our findings establish NEUROeSTIMator as a powerful and broadly applicable tool for measuring neuronal activation, whether as a critical covariate or a primary readout of interest.

AB - Neuronal activity-dependent transcription directs molecular processes that regulate synaptic plasticity, brain circuit development, behavioral adaptation, and long-term memory. Single cell RNA-sequencing technologies (scRNAseq) are rapidly developing and allow for the interrogation of activity-dependent transcription at cellular resolution. Here, we present NEUROeSTIMator, a deep learning model that integrates transcriptomic signals to estimate neuronal activation in a way that we demonstrate is associated with Patch-seq electrophysiological features and that is robust against differences in species, cell type, and brain region. We demonstrate this method’s ability to accurately detect neuronal activity in previously published studies of single cell activity-induced gene expression. Further, we applied our model in a spatial transcriptomic study to identify unique patterns of learning-induced activity across different brain regions in male mice. Altogether, our findings establish NEUROeSTIMator as a powerful and broadly applicable tool for measuring neuronal activation, whether as a critical covariate or a primary readout of interest.

UR - http://www.scopus.com/inward/record.url?scp=85183328270&partnerID=8YFLogxK

U2 - 10.1038/s41467-023-44503-5

DO - 10.1038/s41467-023-44503-5

M3 - Article

SN - 2041-1723

VL - 15

JO - Nature Communications

JF - Nature Communications

IS - 1

M1 - 779

ER -

Using deep learning to quantify neuronal activation from single-cell and spatial transcriptomic data

Abstract

Access to Document

Other files and links

Fingerprint

Cite this