MatNWB Logo

codecov Azure DevOps tests

MatNWB is a Matlab interface for reading and writing Neurodata Without Borders (NWB) 2.x files.


Step 1: Download MatNWB

Download the current release of MatNWB from the MatNWB releases page or from the View NeurodataWithoutBorders/matnwb on File Exchange. You can also check out the latest development version via

git clone

Step 2: Generate the API

From the Matlab command line, add MatNWB to the Matlab path and generate the core classes for the NWB schema.

cd matnwb
generateCore(); % generate the most recent nwb-schema release.

Optional: Generate MatNWB code for extensions

The generateExtension command generates extension classes given a file path to the extension’s namespace.

generateExtension('schema/core/nwb.namespace.yaml', '.../my_extensions1.namespace.yaml',...);

Generated Matlab code will be put into a +types subdirectory. This is a Matlab package. When the +types folder is accessible to the Matlab path, the generated code will be used for reading NWBFiles.

nwb = nwbRead('data.nwb');

Optional: Generate an older core schema

The generateCore command can generate older versions of the nwb schema.


Currently, only schemas >= 2.2.0 are supported (2.1.0 and 2.0.1 are not supported at this time).


Intro to MatNWB

Extracellular Electrophysiology YouTube walkthrough
Calcium Imaging YouTube walkthrough

Intracellular Electrophysiology


Dynamic tables


Advanced data write YouTube walkthrough

Using Dynamically Loaded Filters

Remote read

API Documentation

For more information regarding the MatNWB API or any of the NWB Core types in MatNWB, visit the MatNWB API Documentation pages.

How does it work

NWB files are HDF5 files with data stored according to the Neurodata Without Borders: Neurophysiology (NWB:N) schema. The schema is described in a set of yaml documents. These define the various types and their attributes.

This package provides two functions generateCore and generateExtension that transform the yaml files that describe the schema into Matlab m-files. The generated code defines classes that reflect the types defined in the schema. Object attributes, relationships, and documentation are automatically generated to reflect the schema where possible.

Once the code generation step is done, NWB objects can be read, constructed and written from Matlab.

PyNWB’s cached schemas are also supported: any file written in PyNWB, including those with extensions, will be readable in MATLAB without needing to generate the core API (generateCore) or extensions (generateExtension), as long as the schemas are cached (which is default).


MatNWB is available online at

Data Dimensions

NWB files use the HDF5 format to store data. There are two main differences between the way MATLAB and HDF5 represents dimensions. The first is that HDF5 is C-ordered, which means it stores data is a rows-first pattern, and the MATLAB is F-ordered, storing data in the reverse pattern, with the last dimension of the array stored consecutively. The result is that the data in HDF5 is effectively the transpose of the array in MATLAB. The second difference is that HDF5 can store 1-D arrays, but in MATLAB the lowest dimensionality of an array is 2-D. Due to differences in how MATLAB and HDF5 represent data, the dimensions of datasets are flipped when writing to/from file in MatNWB. This behavior differs depending on whether VectorData use DataPipe objects to contain the data. It’s important to keep in mind the mappings below to make sure is written to and read from file as expected.

without DataPipes

Writing to File

in MatNWB
in HDF5
(M, 1) (M,)
(1, M) (M,)
(P, O, N, M) (M, N, O, P)

Reading from File

in HDF5
in MatNWB
(M,) (M,1)
(M, N, O, P) (P, O, N, M)

NOTE: MATLAB does not support 1D datasets. HDF5 datasets of size (M,) are loaded into MATLAB as datasets of size (M,1). To avoid changes in dimensions when writing to/from file use column vectors for 1D datasets.

with DataPipes

Writing to File

in MatNWB
in HDF5
(M, 1) (1, M)
(1, M) (M, 1)/(M,)**
(P, O, N, M) (M, N, O, P)

** Use scalar as input to ‘maxSize’ argument to write dataset of shape (N,)

Reading from File

in HDF5
in MatNWB
(M, 1) (1, M)
(1, M) (M, 1)
(M,) (M, 1)
(M, N, O, P) (P, O, N, M)


The NWB:N schema is in a state of some evolution. This package assumes a certain set of rules are used to define the schema. As the schema is updated, some of the rules may be changed and these will break this package.

For those planning on using matnwb alongside pynwb, please keep the following in mind:

The master branch in this repository is considered perpetually unstable. If you desire matnwb’s full functionality (full round-trip with nwb data), please consider downloading the more stable releases in the Releases tab. Keep in mind that the Releases are generally only compatible with older versions of pynwb and may not supported newer data types supported by pynwb (such as data references or compound types). Most releases will coincide with nwb-schema releases and contain compatibility with those features.

This package reads and writes NWB:N 2.0 files and does not support older formats.


Basic Data Retrieval | showcases how one would read and process converted NWB file data to display a raster diagram.

Conversion of Real Electrophysiology/Optophysiology Data | converts Electrophysiology/Optophysiology Data recorded from:

Li, Daie, Svoboda, Druckman (2016); Data and simulations related to: Robust neuronal dynamics in premotor cortex during motor planning. Li, Daie, Svoboda, Druckman, Nature.

Analysis examples will be added in the dandi-example-live-scripts repo

Third-party Support

The +contrib folder contains tools for converting from other common data formats/specifications to NWB. Currently supported data types are TDT, MWorks, and Blackrock. We are interested in expanding this section to other data specifications and would greatly value your contribution!


Run the test suite with nwbtest.


  1. “A class definition must be in an “@” directory.”

Make sure that there are no “@” signs anywhere in your full file path. This includes even directories that are not part of the matnwb root path and any “@” signs that are not at the beginning of the directory path.

Alternatively, this issue disappears after MATLAB version 2017b. Installing this version may also resolve these issues. Note that the updates provided with 2017b should also be installed.