Support for Neural Recording Data

Neural data is loaded by the function unpackCaData, called by unpackExperiment when data is first loaded. It can load the following data formats:

• .mat files with output from CNMF-E- Bento will check the following:
  • If the file contains a variable neuron created by CNMF-E, Bento will load neuron.C_raw.
  • Otherwise, if the file contains only one variable (eg neuron_ds) Bento will load (struct).C_raw if it exists.
• other .mat files- if the file contains a single matrix, Bento will load that matrix and assume each row is one neuron.
• .csv files- Bento will assume the first column contains time information, and the 5th column is a fiberphotometry trace.
  • this is the Anderson lab format for some fiberphotometry data, but may not be broadly used.

unpackCaData returns two arguments: rast is a neuron-by-time matrix of activity, and time is a vector of time stamps for each frame (if left empty, Bento uses the sampling rate specified in the Experiment Manager.)

User contribution of code for other data formats is welcome, contact kennedya@caltech.edu if interested.