RasterVis is a D3-based visualization tool for quickly viewing, grouping and summarizing spike rasters for many neurons.
This tool allows you to:
- Generate and change between rasters for many neurons
- Quickly view rasters aligned to experimental trial events.
- Add Gaussian-smoothed peristimulus time kernel density estimates with arbitrary smoothing.
- Group spikes based on experimental factors.
See an example here: https://neurophysvis.github.io/RasterVis/public/
To run RasterVis locally, you need to install the dependencies and set up a local web server.
To install the dependencies and start a webserver, run:
npm install
npm run postinstall
All files must be in the JSON format. Time should be measured in milliseconds.
All data must be placed in the /DATA folder
Your data must follow a specific naming scheme to be rendered by RasterVis.
- Sessions must be named as (subject) + (sessionNumber) e.g. "Bob23"
- Neurons must be named as "(sessionName)_(wireNumber)_(unit_number)" e.g. "Bob23_1_1"
There are three main file types:
- trialInfo.json
- _TrialInfo.json e.g. "Bob23_TrialInfo.json"
- Neuron_.json e.g. "Neuron_Bob23_1_1"
The trialInfo.json contains an array of three objects structured as follows:
|--neurons
|--name: Name of neuron
|--sessionName: Name of session
|--timePeriods
|--name: Name of trial event
|--label: Short label of trial event
|--startID: Beginning of trial event
|--endID: End of trial event
|--color: Color of trial event
|--experimentalFactor
|--name: Name of experimental factor to sort by
|--value: experimental factor key
|--factorType: continuous | categorical | ordinal
The (sessionName)_TrialInfo.json is an array of objects corresponding to each trial. Each trial object contains the properties corresponding to trial events and experimental factors. For example:
{
"trial_id": 129,
"start_time": 0,
"fixation_onset": 143,
"rule_onset": 491,
"stim_onset": 934,
"react_time": 1065,
"reward_time": 1204,
"end_time": 1232,
"isCorrect": "Incorrect",
"isIncluded": "Included",
"Fixation_Break": "No Fixation Break"
}
The Neuron_(neuronName).json contains an object structured as follows:
|--Name
|--Brain_Area
|--Subject
|--File_Name
|--Number_of_Trials
|--Spikes: an array of spike objects corresponding to each trial
|--trial_id
|--spikes: an array of spike times