Skip Navigation Bar
National Library of Medicine Technical BulletinNational Library of Medicine Technical Bulletin

Table of Contents: 2021 MAY–JUNE No. 440


UMLS 2021AA Release Available

Rewolinski J and Wilder V. UMLS 2021AA Release Available. NLM Tech Bull. 2021 May-Jun;(440):e2.

2021 May 03 [posted]

The 2021AA release of the Unified Medical Language System® (UMLS) Knowledge Sources is available for download as of May 3, 2021.

The available downloads are:

  • Full Release (UMLS Metathesaurus, Semantic Network, Specialist Lexicon and Lexical Tools, database load scripts, and MetamorphoSys for customizing your UMLS subset and browsing the data)
  • UMLS Metathesaurus Files (complete Metathesaurus data without any customization)
  • RRF file (most widely used Metathesaurus file)

The SOAP API was discontinued on December 31, 2020. SOAP API users are encouraged to switch to the REST API.

Metathesaurus

The 2021AA Metathesaurus contains approximately 4.44 million concepts and 16.1 million unique concept names from 218 source vocabularies.

Sibling relationships were removed from the relationships file (MRREL) in the UMLS Metathesaurus for the May 2021AA release. This change reduced the size of the MRREL file by 40% and the overall size of the Metathesaurus by 10%. Sibling relationships can be computed using the Metathesaurus MRHIER.RRF file. To find an atom's siblings in a specific context, find all MRHIER.RRF rows that share its SAB and PTR values. Learn more about the MRHIER file.

One new MEDLINEPLUS translation:

  • MEDLINEPLUS_SPA is the Spanish Edition of MedlinePlus Health Topics (MEDLINEPLUS)

Three new NCI sub-sources:

  • NCI_CTDC (Clinical Trial Data Commons)
  • NCI_PCDC (Pediatric Cancer Data Commons)
  • NCI_caDSR (Cancer Data Standards Registry and Repository)

Three source abbreviation (SAB) updates:

  • CCSR_10 is now CCSR_ICD10CM (Clinical Classifications Software Refined for ICD-10-CM)
  • CCS_10 is now CCSR_ICD10PCS (Clinical Classifications Software Refined for ICD-10-PCS)
  • NUCCPT is now NUCCHCPT (National Uniform Claim Committee - Health Care Provider Taxonomy)

One LNC sub-source removed:

  • LNC_SPECIAL_USE (LOINC Prerelease Terms, Special Use)

74 English sources and 33 translation sources were updated. These include MeSH®, MedDRA, RxNorm, and SNOMED CT® (English and Spanish).  A complete list is available in the Updated Sources section of the Release Documentation. For more detailed information on changes in this version of the Metathesaurus, see the Updated Sources (Expanded) section. Additional release statistics may be found in the Statistics section.

SPECIALIST Lexicon and Lexical Tools

  • The release includes the updated SPECIALIST Lexicon (2021 Release).
  • The release includes the updated Lexical Tools (2021 Release) which integrate data from the SPECIALIST Lexicon, 2021 Release. The Lexical Tools include the Full and Lite versions of lvg.2021.
  • The Metathesaurus index files were processed using the updated lvg files.
  • The updated SPECIALIST Lexicon and Lexical Tools feature synonym enhancements that include changes to their file structure. See the SPECIALIST Lexicon and Lexical Tools websites for details about these file changes and other updates.

MetamorphoSys

  • The full release requires 33.4 GB of disk space.

Reported bugs may be viewed on the Release Notes and Bugs webpage.

MetamorphoSys can generate custom load scripts for MySQL, Oracle, or Microsoft Access when creating a Metathesaurus subset or installing the Semantic Network. Instructions are available on the UMLS Load Scripts homepage.

Release Information

To access the UMLS Release files, you must have an active UMLS Metathesaurus® License and a valid UTS account. You will be prompted to authenticate with an identity provider with the UTS when downloading the files.

UMLS Learning Resources
2021AA Source Release Documentation webpages are available. 

Additional information regarding the UMLS is available on the UMLS homepage. New users are encouraged to take the UMLS Basics Tutorial and to explore the UMLS Quick Start Guide, and other training materials.

UMLS Terminology Services (UTS)

The UTS Metathesaurus Browser and the UTS REST API include the updated release.

UMLS User Contributions

UMLS users have extended the functionality of the UMLS in a variety of ways by developing APIs, automation scripts, and natural language processing tools. You can find a list of these on the UMLS Community webpage.

Want to add your tool? Send a request to the NLM Help Desk with the subject line: "UMLS Community."

Be sure to include a link to your source code so that other UMLS users can adapt your tool. We are especially interested in:

  • Database load scripts
  • Transformation scripts that convert UMLS data into other formats (for example, RDF or JSON)
  • Scripts that automate any aspect of UMLS installation
  • Applications that leverage UMLS in the processing of text

By Jenny Rewolinski and Victoria Wilder
MEDLARS Management Section

NLM Technical Bulletin National Library of Medicine National Institutes of Health