Single cortical neurons as deep artificial neural networks

This dataset contains training and testing subsets of the input/output relationship of a single cortical layer 5 pyramidal cell (L5PC) neuron at 1ms single spike temporal resolution.
The data is obtained via a simulation that contains all of the currently (2021) known and well modeled "messy biological details" that relate to the operation of single neurons in the brain.

The goal with this dataset is to allow machine learning modeling experts easier access to high quality biological data and eventually find as-small-as-possible models that as-accurately-as-possible capture the simulation data of a single cortical neuron at 1ms temporal resolution. Where in this case "a small model" can refer to: "fast", "parameter efficient", "conceptually simple", "elegant", etc.

All related resources

Github repo: github.com/SelfishGene/neuron_as_deep_net
Neuron version of paper: cell.com/neuron/fulltext/S0896-6273(21)00501-8
Open Access (slightly older) bioRxiv version of Paper: biorxiv.org/content/10.1101/613141v2
Dataset and pretrained networks: kaggle.com/selfishgene/single-neurons-as-deep-nets-nmda-test-data
Dataset for training new models: kaggle.com/selfishgene/single-neurons-as-deep-nets-nmda-train-data
Notebook with main result: kaggle.com/selfishgene/single-neuron-as-deep-net-replicating-key-result
Notebook exploring the dataset: kaggle.com/selfishgene/exploring-a-single-cortical-neuron
Twitter thread for short visual summery #1: twitter.com/DavidBeniaguev/status/1131890349578829825
Twitter thread for short visual summery #2: twitter.com/DavidBeniaguev/status/1426172692479287299
Figure360, author presentation of Figure 2 from the paper: youtube.com/watch?v=n2xaUjdX03g

If you use this dataset or associated models or code, please cite the following two works:

1. David Beniaguev, Idan Segev and Michael London. "Single cortical neurons as deep artificial neural networks." Neuron. 2021; 109: 2727-2739.e3 doi: https://doi.org/10.1016/j.neuron.2021.07.002
2. Hay, Etay, Sean Hill, Felix Schürmann, Henry Markram, and Idan Segev. 2011. “Models of Neocortical Layer 5b Pyramidal Cells Capturing a Wide Range of Dendritic and Perisomatic Active Properties.” Edited by Lyle J. Graham. PLoS Computational Biology 7 (7): e1002107. doi: https://doi.org/10.1371/journal.pcbi.1002107.