HGVS Nomenclature

Key facts

Maintained by

HGVS Variant Nomenclature Committee (HVNC), under HUGO

Reference prefixes

g. genomic, c. coding DNA, n. non-coding, m. mitochondrial, r. RNA, p. protein

Clinical transcript set

MANE Select, from NCBI and EMBL-EBI

What is HGVS nomenclature?

HGVS nomenclature gives every sequence variant an unambiguous written description. It was developed through the Human Genome Variation Society, which gives it its name, and is now administered by the HGVS Variant Nomenclature Committee under the Human Genome Organization (HUGO).

A complete HGVS description has two parts: an identifier for the reference sequence, including its version, and a description of the change relative to that sequence. Without the reference, the same position number can point to different bases.

Every character carries meaning. The prefix letter, the period, the reference version and the change symbol together identify one variant, so reformatting a description by hand can silently change or break it.

How an HGVS description is structured

A familiar oncology example is a substitution in the BRAF gene written c.1799T>A at the coding DNA level, which predicts the protein change p.Val600Glu, often shortened to V600E. Key conventions include:

• Reference sequences are usually NCBI RefSeq accessions: NC_ for chromosomes, NM_ for transcripts, NP_ for proteins.
• In c. numbering, position 1 is the A of the ATG start codon; intronic positions are written relative to the nearest exon, as in c.123+1G>A.
• Change types use short codes: > for substitution, del, dup, ins and delins.
• Protein changes predicted from DNA, rather than measured, are placed in parentheses, as in p.(Val600Glu).
• Genomic descriptions on a chromosome accession do not depend on transcript choice, which makes them the most stable form for matching across sources.

Why HGVS nomenclature matters

Variant data from laboratories, publications, drug labels and ClinVar only line up when they describe variants the same way. Protein shorthand such as V600E is convenient but can map to more than one DNA change, so joins built on protein strings alone can merge distinct variants.

The choice of transcript also changes c. and p. numbering. The MANE Select transcript set gives one agreed transcript per gene for clinical reporting, which reduces mismatches when combining sources.