There is a newer version of the record available.

Published February 9, 2026 | Version v40
Dataset Open

PMDB: a relational database for PubMed

  • 1. Vanderbilt University Medical Center

Description

The files constitute a compressed dump of PMDB, which was created in PostgreSQL 17 using the pmparser R package. Once you have a Postgres server running, you can set up the database as follows:

1. Untar the file containing the database dump, which will create a folder. Substitute <tar_name> with the name of the tar file.

tar xvf <tar_name> && rm <tar_name>

2. Restore the database onto your Postgres server. Below is one way. Replace <...> as appropriate, substituting <dump_name> with the name of the new folder containing the dump files. This will take some time, so use as many cores as you can spare. Beware the full database is about 180 GB.

createdb -h <host_name> -U <user_name> pmdb
pg_restore -j <num_cores> --no-privileges --no-owner -h <host_name> -U <user_name> -d pmdb <dump_name>

MEDLINE/PubMed data are courtesy of the U.S. National Library of Medicine. See NLM's Terms and Conditions.

Files

Files (33.0 GB)

Name Size Download all
md5:67215b46cbff0ed99d80f20338e7f787
33.0 GB Download

Additional details

Related works

Is referenced by
Preprint: 10.1101/2020.09.07.285924 (DOI)