Skip to content

Latest commit

 

History

History
98 lines (68 loc) · 5.7 KB

File metadata and controls

98 lines (68 loc) · 5.7 KB

replay_trajectory_classification

DOI PyPI version Binder PR Test

Installation | Documentation | Tutorials | References | Developer Installation

What is replay_trajectory_classification?

replay_trajectory_classification is a python package for decoding spatial position represented by neural activity and categorizing the type of trajectory.

Advantages over other algorithms

It has several advantages over decoders typically used to characterize hippocampal data:

  1. It allows for moment-by-moment estimation of position using small temporal time bins which allow for rapid movement of neural position and makes fewer assumptions about what downstream cells can integrate.
  2. The decoded trajectories can change direction and are not restricted to constant velocity trajectories.
  3. The decoder can use spikes from spike-sorted cells or use clusterless spikes and their associated waveform features to decode .
  4. The decoder can categorize the type of neural trajectory and give an estimate of the confidence of the model in the type of trajectory.
  5. Proper handling of complex 1D linearized environments
  6. Ability to extract and decode 2D environments
  7. Easily installable, documented code with tutorials on how to use the code (see below)
  8. Fast computation using GPUs. (Note: must install cupy to use)

References

For further details, please see our eLife paper:

Denovellis, E.L., Gillespie, A.K., Coulter, M.E., Sosa, M., Chung, J.E., Eden, U.T., and Frank, L.M. (2021). Hippocampal replay of experience at real-world speeds. ELife 10, e64505.

or our conference paper:

Denovellis, E.L., Frank, L.M., and Eden, U.T. (2019). Characterizing hippocampal replay using hybrid point process state space models. In 2019 53rd Asilomar Conference on Signals, Systems, and Computers, (Pacific Grove, CA, USA: IEEE), pp. 245–249.

Also see other work using this code:

Gillespie, A.K., Astudillo Maya, D.A., Denovellis, E.L., Liu, D.F., Kastner, D.B., Coulter, M.E., Roumis, D.K., Eden, U.T., and Frank, L.M. (2021). Hippocampal replay reflects specific past experiences rather than a plan for subsequent choice. Neuron S0896627321005730. https://doi.org/10.1016/j.neuron.2021.07.029.

Joshi, A., Denovellis, E.L., Mankili, A., Meneksedag, Y., Davidson, T., Gillespie, K., Guidera, J.A., Roumis, D., and Frank, L.M. (2022). Dynamic Synchronization between Hippocampal Spatial Representations and the Stepping Rhythm. bioRxiv, 30. https://doi.org/10.1101/2022.02.23.481357.

Gillespie, A.K., Astudillo Maya, D.A., Denovellis, E.L., Desse, S., and Frank, L.M. (2022). Neurofeedback training can modulate task-relevant memory replay in rats. bioRxiv, 2022.10.13.512183. https://doi.org/10.1101/2022.10.13.512183.

Installation

replay_trajectory_classification can be installed through pypi or conda. Conda is the best way to ensure that all the dependencies are installed properly.

pip install replay_trajectory_classification

Or

conda install -c edeno replay_trajectory_classification

Documentation

Documentation can be found here: https://replay-trajectory-classification.readthedocs.io/en/latest/

Tutorials

There are five jupyter notebooks introducing the package:

  1. 01-Introduction_and_Data_Format: How to get your data in the correct format to use with the decoder.
  2. 02-Decoding_with_Sorted_Spikes: How to decode using a single movement model using sorted spikes.
  3. 03-Decoding_with_Clusterless_Spikes: How to decode using a single movement model using the "clusterless" approach --- which does not require spike sorting.
  4. 04-Classifying_with_Sorted_Spikes: Using multiple movement models to classify the movement dynamics and decode the trajectory using sorted spikes.
  5. 05-Classifying_with_Clusterless_Spikes: Using multiple movement models to classify the movement dynamics and decode the trajectory using clusterless spikes.

Developer Installation

For people who want to expand upon the code for their own use:

  1. Install miniconda (or anaconda) if it isn't already installed. Type into bash (or install from the anaconda website):
wget https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh -O miniconda.sh;
bash miniconda.sh -b -p $HOME/miniconda
export PATH="$HOME/miniconda/bin:$PATH"
hash -r
  1. Go to the local repository on your computer (cd replay_trajectory_classification) and install the anaconda environment for the repository. Type into bash:
conda update -n base conda # make sure conda is up to date
conda env create -f environment.yml # create a conda environment
conda activate replay_trajectory_classification # activate conda environment
python setup.py develop