paper-ephys-atlas

Code related to the electrophysiology atlas paper

Getting started

To get the features dataframe needed to train models, see the loading example

To get a naive example of decoding model out of ephys features, see a raw ephys decoder

Configuration instructions

The configuration file is located in config-ephys-atlas.yaml. It contains local data storage paths and some parameters for the analysis. To avoid committing changes to this file, it is ignored by git. To make changes to the configuration, copy the template file and edit the copy:

cp config-ephys-atlas.template.yaml config-ephys-atlas.yaml

Install instructions

Clone the repository

git clone https://github.com/int-brain-lab/paper-ephys-atlas.git

Activate your environment of choice (usually a conda iblenv as described here: https://github.com/int-brain-lab/iblenv).

conda activate iblenv

Cd into the repository and install in-place

cd paper-ephys-atlas
pip install -e .

Download the features table

The features table is stored with our aggregates dataset as part of IBL data. To download it, run the following commands in Python:

from pathlib import Path
from one.api import ONE
import ephys_atlas.data

LOCAL_DATA_PATH = Path.home().joinpath("Downloads")
LABEL = "2023_W41"  # or put "latest"
one = ONE(base_url="https://alyx.internationalbrainlab.org", mode='local')
df_raw_features, df_clusters, df_channels, df_probes = ephys_atlas.data.download_tables(label=LABEL, local_path=LOCAL_DATA_PATH, one=one)