PDB structures in the MMTF format are updated weekly on Wednesday around 00:00 UTC.

Atomic Level of Detail and Data Precision

MMTF data are available in two representations:
All atom representation
0.001Å coordinate precision, 0.01 B-factor and occupancy precision
C-alpha atoms only for polypeptides
P-backbone atoms only for polynucleotides
All atom representation for all other residue types
0.1Å coordinate precision, 0.1 B-factor and occupancy precision.
The reduced representation is up to a factor of 10 smaller than the full representation. It is useful for the efficient visualization of ribbon diagrams or for calculations that only use C-alpha atoms.


The version number is embedded in the base URLs{version}

For example, points to API version v1.0.*. Minor versions will be backward compatibile. For changes in the major version number you need to update your code to point to the new version of the API.

The current major version of the API is v1

RESTful Services

The RESTful service returns a response as a MessagePack container with MMTF data.

REST URLs for full and reduced representation using either HTTP or HTTPS:{PDB ID} and{PDB ID} or{PDB ID} and{PDB ID}

For example to access the PDB structure 4HHB in the all atom (full) representation use:


MMTF Files

Individual MMTF files can be downloaded in gzipped format with command line utilties such as wget or curl:
$ wget
$ curl -O

Hadoop Sequence File

Downloading and processing many individual files is inefficient. For large scale calculations with distributed parallel frameworks such as Apache Hadoop or Apache Spark we recommend the use of Hadoop Sequence Files. The entire PDB archive can be downloaded in the full and reduced representations: and

$ wget
$ tar -xvf full.tar
$ curl -O
$ tar -xvf full.tar

This will download and unpack the content of the Hadoop Sequence File to a folder.