Hester M. Wain, Elspeth A. Bruford, Ruth C. Lovering, Michael J. Lush, Mathew W. Wright and Sue Povey
Genomics 79(4):464-470 (2002)
1. Criteria for Symbol Assignment
2. Gene Symbols
3. Gene Names
4. Gene Families
5. Homologies with Other Species
6. Genes Identified from Sequence Information
7. Enzymes and Proteins
8. Clinical Disorders
9. Genomic Rearrangements and Features
10. Characters Reserved for Specific Usage
11. Symbol Status
Appendix 1: Gene Symbol Use in Publications
Appendix 2: Tables
Table 1: Greek-to-Latin Alphabet Conversion
Table 2: Species Abbreviations
Table 3: Single-letter Amino Acid Symbols
Please note that not all browsers support the fonts used in this document, this is particularly pertinent when viewing the Greek-Latin letter conversion table.
Guidelines for human gene nomenclature were first published in 1979 , when the Human Gene Nomenclature Committee was first given the authority to approve and implement human gene names and symbols. Updates of these guidelines were published in 1987 ,1995 , and 1997 . With the recent publications of the complete human genome sequence there is an estimated total of 26,000-40,000 genes, as suggested by the International Human Genome Sequencing Consortium  and Venter et al. . Thus, the guidelines have been updated to accommodate their application to this wealth of information, although symbols are still only assigned when required for communication. These updates were derived with input from the HUGO Gene Nomenclature Committee (HGNC) International Advisory Committee and attendees of the ASHG01NW Gene Nomenclature Workshop. All approved human gene symbols can be found in the HGNC database .
The philosophy of the HGNC remains "that gene nomenclature should evolve with new technology rather than be restrictive as sometimes occurs when historical and single gene nomenclature systems are applied" .
A summary of the guidelines is presented here:
1. Each approved gene symbol must be unique.
2. Symbols are short-form representations (or abbreviations) of the descriptive gene name.
3. Symbols should only contain Latin letters and Arabic numerals.
4. Symbols should not contain punctuation.
5. Symbols should not contain "G" for gene.
6. Symbols do not contain any reference to species, for example "H/h" for human.
A gene is defined as: "a DNA segment that contributes to phenotype/function. In the absence of demonstrated function a gene may be characterized by sequence, transcription or homology". The overwhelming majority of objects named by HGNC are in this category.
The word "locus" is not a synonym for gene but refers to a map position. A more precise definition is given in the Rules and Guidelines from the International Committee on Standardized Genetic Nomenclature for Mice which states: "A locus is a point in the genome, identified by a marker, which can be mapped by some means. It does not necessarily correspond to a gene; it could, for example, be an anonymous non-coding DNA segment or a cytogenetic feature. A single gene may have several loci within it (each defined by different markers) and these markers may be separated in genetic or physical mapping experiments. In such cases, it is useful to define these different loci, but normally the gene name should be used to designate the gene itself, as this usually will convey the most information."
1.3 Chromosome Region
In the context of gene nomenclature "chromosome region" is defined as a genomic region which has been associated with a particular syndrome or phenotype, particularly when there is a possibility that several genes within it may be involved in the phenotype. Designation of such regions may be requested by the scientific community and approved by HGNC e.g. ANCR "Angelman syndrome chromosome region" and CECR "cat eye syndrome chromosome region".
Symbols may therefore be assigned to the following:
a) Clearly defined phenotypes shown to be inherited predominantly as monogenic Mendelian traits e.g. BBS1 "Bardet-Biedl syndrome 1".
b) Unidentified genes contributing to a complex trait shown by linkage, or association with a known marker e.g. IDDM6 "insulin-dependent diabetes mellitus 6".
c) Cloned segments of DNA with sufficient structural, functional, and expression data to identify them as transcribed entities e.g. COX8A "cytochrome c oxidase subunit VIIIA".
d) Non-functional copies of genes (pseudogenes) e.g. IL9RP1 "interleukin 9 receptor pseudogene 1".
e) Genes encoded by the opposite (antisense) strand that overlap a known gene e.g. IGF2-AS "IGF2 antisense RNA".
f) Transcribed but untranslated functional DNA segments e.g. XIST "X (inactive)-specific transcript".
g) Cellular phenotypes from which the existence of a gene or genes can be inferred e.g. LOH18CR1 "loss of heterozygosity, 18, chromosomal region 1".
h) EST clusters which suggest a putative gene e.g. C3orf20 "chromosome 3 open reading frame 20".
i) Fragments of expressed sequence will be designated a D-number by GDB (the Genome Database) e.g. DXYS155E (Appendix 1)
j) Polycistronic genes generated from a single mRNA, but with independent coding sequence, physically separable and non-overlapping with other coding sequence giving independent gene products e.g. SNURF "SNRPN upstream reading frame" and SNRPN "small nuclear ribonucleoprotein polypeptide N".
k) Genes of unknown function which share highly similar sequences e.g. FAM9A "family with sequence similarity 9, member A" etc.
l) Predicted genes (in silico) which show a high degree of sequence homology to well characterized genes will be assigned the same symbol with an "L" for like e.g. TCP10L "t-complex 10 (mouse)-like".
m) Intronic transcripts (on the same DNA strand) will be assigned separate symbols, usually relating to the gene in which they reside e.g. COPG2IT1 "coatomer protein complex, subunit gamma 2, intronic transcript 1".
n) Transcriptional units that are constituents of a named complex locus, provided the subsidiary nomenclature is required for clarity of scientific communication. e.g. "UGT1A1", a constituent of the "UGT1A" complex locus.
Gene symbols will not usually be assigned to alternative transcripts or to genes predicted solely from in silico data (with no other supporting evidence e.g. significant homology to a characterized gene).
Human gene symbols are designated by upper-case Latin letters or by a combination of upper-case letters and Arabic numerals, with the exception of the C#orf#symbols. Symbols should be short in order to be useful, and should not attempt to represent all known information about a gene . Symbols should be inoffensive and should not spell words or match abbreviations that would cause problems with database searching e.g. DNA. Ideally, symbols should be no longer than six characters in length. New symbols must not duplicate existing approved gene symbols in either the human (HGNC database) or the mouse (MGD) databases.
a) The initial character of the symbol should always be a letter. Subsequent characters may be other letters, or if necessary, Arabic numerals.
b) All characters of the symbol should be written on the same line; no superscripts or subscripts may be used.
c) No Roman numerals may be used. Roman numbers in previously used symbols should be changed to their Arabic equivalents.
d) Greek letters are not used in gene symbols. All Greek letters should be changed to letters in the Latin alphabet (App 2: Table 1). Note that such gene symbols will then appear in lists in the order of the Latin alphabet.
e) A Greek letter prefixing a gene name must be changed to its Latin alphabet equivalent and placed at the end of the gene symbol. This permits alphabetical ordering of the gene in listings with similar properties, such as substrate specificities e.g. GLA "galactosidase, alpha"; GLB "galactosidase, beta".
f) No punctuation may be used, with the exception of hyphens in specific families and locus types, such as the HLA, immunoglobulin and T cell receptor gene symbols (which may be hyphenated). The HLA symbols are assigned by the WHO Nomenclature Committee for Factors of the HLA System  via the IMGT/HLA database . The immunoglobulin and T cell receptor gene symbols are assigned by the IMGT Nomenclature Committee via the IMGT/LIGM database .
g) Gene symbols will not usually be assigned to alternative transcripts. However, if a community working on a group of genes has a need for nomenclature where there are multiple small coding sequences which can be combined to form a number of different larger products then these coding sequences may be assigned symbols e.g. UGT1A1 "UDP glycosyltransferase 1 family, polypeptide A1" to UGT1A13 representing 13 distinct gene symbols.
h) Reference to tissue specificity, molecular weight or chromosomal location should be avoided, where necessary this may be included in the gene name.
i) Some letters or combination of letters are used as prefixes or suffixes in a symbol to give a specific meaning and their use for other meanings should be avoided (Section 10).
j) Oncogenes are given symbols corresponding to the homologous retroviral oncogene, but without the "v-" or "c-" prefices e.g. JUN "v-jun sarcoma virus 17 oncogene homolog (avian)", and SRC "v-src sarcoma (Schmidt-Ruppin A-2) viral oncogene homolog (avian)".
Gene names should be brief and specific and should convey the character or function of the gene, but should not attempt to describe everything known about it. The first letter of the symbol should be the same as that of the name in order to facilitate alphabetical listing and grouping. Gene names are written using American spelling. Tissue specificity and molecular weight designations should be avoided as they have only limited use as a description and may in time and across species prove inaccurate; however, they may be incorporated into the gene name if absolutely necessary. If necessary, molecular weights may be specified in kilodaltons using the SI unit: kDa with no space after the molecular weight. Gene names should not include terms such as nephew, cousin, sister etc. to describe familial relationships with other genes. The following gene name syntax should be used:
a) Names start with a lower case letter unless it is a person's name describing a disease/phenotype or a capitalised abbreviation e.g. AHDS "Allan-Herndon-Dudley syndrome" and ABCA1 "ATP-binding cassette, sub-family A (ABC1), member 1", respectively.
b) Descriptive modifiers should follow the main part of the name, separated by commas e.g. ACOT1 "acyl-CoA thioesterase 1".
c) Where a complete alternative name (or names) is being included as part of the name, this should be in parentheses e.g. IDS "iduronate 2-sulfatase (Hunter syndrome)".
d) Names of other species must be placed in parentheses at the end e.g. LFNG "lunatic fringe homolog (Drosophila)" and ANLN "anillin, actin binding protein (scraps homolog, Drosophila)".
Hierarchical symbols for both structural and functional gene families should be used where possible. A stem (or root) symbol as a basis for a symbol series allows easy identification of other family members in both database searches and the literature. Gene family members should be designated by Arabic numerals placed immediately after the gene stem symbol, without any space between the letters and numbers used e.g. GPR1, GPR2, GPR3 (three G protein-coupled receptor genes). However, very occasionally, only if they exist historically, single-letter suffixes may be used to designate these different genes e.g. LDHA, LDHB, LDHC (three lactate dehydrogenase genes). Some gene families are very large and so may include a variety of number/letter combinations to indicate their phylogenetic relationships e.g. CYP1A1, CYP21A2, CYP51A1 (three members of the cytochrome P450 superfamily). When symbols are approved by the HGNC consecutively in a gene family the hierarchical order of symbols will not necessarily reflect the chronological order of peer-reviewed publications. Consecutive symbols approved by HGNC take precedence over those published, although this will be a matter for discussion with the relevant scientific community.
Many genes receive approved symbols and names which are non-ideal when considered in the light of subsequent information. In the case of individual genes a change to the name (and subsequently the symbol) is only made if the original name is seriously misleading. However, groups of scientists working on particular gene families often co-ordinate a revised nomenclature when more information becomes available; such initiatives are welcomed. Groups planning to do this are strongly advised to liase with the HGNC in order to avoid possible problems. Previously approved symbols which have been withdrawn should not be re-used where possible, as this causes great confusion in information retrieval.
4.1 Anonymous families
When a series of genes can be shown to be related by sequence similarity but otherwise cannot be described by homology or function they can be assigned anonymous and temporary FAM# symbols. Membership in each family, indicated by the FAM number, will be assigned using the general criteria established by Mackenzie et al.  and Nelson et al.  that the gene products show greater than 40% amino acid identity. However, these criteria are not exclusive but merely a convenient starting point on which to base an appropriate symbol e.g. FAM9A, FAM9B and FAM9C. Once further information is available to assign a more descriptive name the whole gene family should be updated with new symbols; the FAM designations being withdrawn.
Homologous genes in different vertebrate (and where appropriate, non-vertebrate) species (orthologs) should where possible have the same gene nomenclature.
a) Human homologs of genes first identified in other species should not be designated by a symbol beginning with H or h for human.
b) When a gene or series of genes has been defined in one species, and it is reasonable to expect that in the future a homologous gene will be identified in man, we recommend that the designated symbol be reserved for the human locus. We recommend that this should be done in other species for genes first identified in human.
c) When necessary to distinguish the species of origin for homologous genes with the same gene symbol, the letter-based code for different species already established by SWISS-PROT should be used. This can be found at URL <http://www.expasy.ch/cgi-bin/speclist> and commonly used species are shown in App 2: Table 2. The codes listed in Table 2 are no longer provided by SWISS-PROT, and may have been replaced by those listed in the UniProt species list (see http://www.uniprot.org/docs/speclist). In light of this change our recommendation of which species codes to use is currently under review.
d) The agreement between human and mouse gene nomenclature for many homologous genes should be continued and extended to other vertebrate species where possible.
e) Human homologs of genes in invertebrate, or prokaryote species may sometimes be represented by the symbol used in the other species, sometimes followed by an L to represent "like", and a number or letter if there is (or is likely to be) more than one human homolog and the organism designated at the end of the gene name in parentheses e.g. FEM1A "fem-1 homolog a (C. elegans)" and ALKBH1 "alkB, alkylation repair homolog 1 (E. coli)". The use of H to represent homolog is no longer recommended, and will be discontinued.
f) For orthologs of human C#orf# genes we suggest following the same nomenclature principle used in the mouse assignment of human D# segment numbers. Mouse DNA segments that are detected by cross-hybridization to human segments are assigned the human symbol with the mouse chromosome number as a prefix and an H denoting human. Thus human D21S56 becomes mouse D16H21S56. Therefore, if C6orf19 in the human is on chicken chromosome 16, the chicken symbol would be C16H6orf19 with the corresponding name "chromosome 16 C6orf19 homolog".
Protein-coding genes of unknown function on the opposite strand, which have no proven regulatory function, should be assigned the suffix OS for "opposite strand".
Non-protein coding RNA genes encode functional RNA molecules that are not translated into proteins. Genes encoding small (<200nt) non-coding RNAs are named after the specific class of RNA product as identified by sequence similarity, e.g. SNORA1, "small nucleolar RNA, H/ACA box 1"; TRNAA1, "transfer RNA alanine 1 (anticodon UGC)". For a review see: PMID:21296742 PMCID:PMC3051107)
Genes encoding long (>200nt) non-coding RNAs (lncRNAs) are preferably named based on the known function of the transcript, e.g. XIST. Antisense transcripts are named using the approved HGNC symbol for the protein-coding gene with the suffix 'AS' for 'antisense' (the gene symbol should not be the protein-coding gene symbol written backwards), intronic transcripts are appended with 'IT', and overlapping transcripts with 'OT'. Intergenic lncRNA genes are provided with a LINC name for 'long intergenic non-protein coding RNA, e.g. LINC00028. A short guide to lncRNA gene nomenclature was published in 2014, see http://www.humgenomics.com/content/8/1/7)
6.3 Related (-like) sequences
The designation of the suffix "L" for like has been used in the past for related sequences identified by cross-hybridization studies. Where genes are identified by database searching and where no other functional information is available and there is some sequence similarity with a known gene they are designated with the symbol of the known gene followed by an "L" for like e.g. ADAL "adenosine deaminase-like". Alternatively, if there are a number of similar genes they may be assigned numbers in a series e.g. OSBPL1A "oxysterol binding protein-like 1A" to OSBPL11 "oxysterol binding protein-like 11".
Genes predicted from EST clusters or from genomic sequence with EST evidence, but showing no structural or functional homology, are regarded as putative. These are designated by the chromosome of origin, the letters "orf" for open reading frame and a number in a series e.g. C3orf17, "chromosome 3 open reading frame 17". The use of the lower case letters "orf" is to prevent confusion between the first letter "o" and the numeral "0" (zero), which may be part of the chromosome number. KIAA# identifiers will be approved as temporary symbols only for those genes mentioned in publications when no other information is known about the gene. The complete list of KIAA# identifiers can be found at the Human cDNA project at the Kazusa DNA Research Institute. It is intended that these temporary symbols will be changed when function is identified.
Pseudogenes are sequences that are generally untranscribed and untranslated and which have high homology to identified genes . However, it has recently been shown that in different organisms or tissues functional activation may occur. Therefore, the previous policy of assigning the gene symbol of the structural gene followed by "P" and a number will only be approved on a case by case basis. In future, pseudogenes will usually be assigned the next number in the relevant symbol series, suffixed by a "P" for pseudogene (or "PS" in the specific cases) if requested e.g. OR5B12P "olfactory receptor, family 5, subfamily B, member 12 pseudogene". However, the designation "pseudogene" will remain in the gene name.
Potential pseudogenes identified from sequence information will usually be assigned a symbol if they show a minimum of 50% predicted amino acid identity across 50% of the open reading frame, to the named gene (adapted from the Human Annotation Workshop 2 definition).
Note that no orthologous relationships can be inferred from the numbering of pseudogenes in other species, as this numbering is usually species-specific.
The rules described in the sections on gene names and symbols apply, but in addition the names of genes coding for enzymes should be based on those recommended by the Nomenclature Committee of the International Union of Biochemistry and Molecular Biology e.g. FPGS "folylpolyglutamate synthase". These can be found at URL <http://expasy.org/enzyme/>. Names of genes encoding plasma proteins, hemoglobins, and specialized proteins are based on standard names and those recommended by their respective committees e.g. HBA1 "hemoglobin, alpha 1".
The first gene symbol allocated to an inherited clinical phenotype (monogenic Mendelian inheritance) may be based on an acronym which has been established as a name for the disorder, whilst following the rules described previously e.g. ACH for "achondroplasia". It is usual for this symbol to change when the gene product or function is identified; however, if there is no additional information derived from the cloned gene, the disease symbol e.g. ACH will be maintained. If an approved gene symbol for the cloned gene based on product or function already exists this will take precedence over the symbol derived from the clinical disorder when the gene descriptions are merged. For example, in the case of achondroplasia the symbol ACH was withdrawn and FGFR3 and the name "fibroblast growth factor receptor 3 (achondroplasia, thanatophoric dwarfism)" approved in its place. If the original symbol is to be maintained we do not recommend the sole use of invented names based upon the phenotype e.g. tuberin for TSC2 "tuberous sclerosis 2".
8.1 Complex/polygenic traits
Genome searches may suggest a contributing locus in a complex trait, which may for convenience be given a gene symbol e.g. IDDM3 "insulin-dependent diabetes mellitus 3", although a proportion of these will disappear intime.
8.2 Contiguous gene syndromes
Syndromes clearly associated with multiple widely dispersed loci should not be given gene symbols. Syndromes associated with a regional deletion or duplication may be assigned the letters CR (for chromosome region), in place of S for syndrome. Examples include ANCR "Angelman syndrome chromosome region", CECR "cat eye syndrome chromosome region". However, as advances in database design have increased the possible ways of representing this type of information, we recommend that such symbols are classified as syndromic region symbols and not gene symbols. Candidate genes found within a chromosomal region may be assigned a symbol based on the syndromic region symbol with sequential numbering e.g. CECR1 "cat eye syndrome chromosome region, candidate 1".
8.3 Loss of heterozygosity
A chromosomal region in which the existence of genes may be inferred by loss of heterozygosity can be designated by a symbol consisting of the letters LOH for loss of heterozygosity, the chromosome number, CR (for chromosomal region), and then a sequential number e.g. LOH1CR1 "loss of heterozygosity, 1, chromosomal region 1".
Symbols will be assigned to genes generated by recombination events. The symbols will be designated a number or letter in the series determined from the original gene symbol e.g. KIR2DL5 gene is duplicated in some haplotypes, the original gene will be referred to as KIR2DL5A "killer cell immunoglobulin-like receptor, two domains, long cytoplasmic tail, 5A" and the duplication as KIR2DL5B "killer cell immunoglobulin-like receptor, two domains, long cytoplasmic tail, 5B". Information detailing position on the chromosome e.g. telomeric may be added to the gene name to aid understanding. These types of symbols will be agreed on an individual basis.
9.2 Gene recombination
When there has been recombination between genes, generating a new locus, the gene symbol should remain short but reflect the two (or more) initial gene symbols. The name may be as descriptive as necessary e.g. POMZP3 "POM (POM121 homolog, rat) and ZP3 fusion".
Letters, or combinations of letters are used as prefixes or suffices in a symbol to represent a specific meaning and so their use for other meanings should be avoided where possible (Table 1). The following letter usage is not acceptable; H or h for human, G or g for gene. If the name of a gene contains a character or property for which there is a recognized abbreviation, the abbreviation should be used e.g. the single-letter abbreviation for amino acids used in aminoacyl residues (App 2: Table 3). Other characters are also used within the symbol to represent genomic or genetic information (Table 1).
Table 1: Characters reserved for specific usage
|C#orf#||chromosome # open reading frame #|
|D or DC||domain containing|
|FAM||family with sequence similarity|
|HG|| host gene
|IN or N||inhibitor|
|LOH||loss of heterozygosity|
|MT or M||mitochondrial|
|@||gene cluster in chromosomal region|
11. Symbol Status
Each gene record in the HGNC database has one of five states:
a) Pending: gene symbols in progress.
b) Approved: official gene symbols that are publicly available.
c) Reserved : official gene symbols that are not publicly available, these will be released after a maximum period of six months.
d) Symbol Withdrawn: previous symbols for genes which now have different approved symbols.
e) Entry Withdrawn: for symbols which refer to a gene that has since been shown not to exist.
It is recommended that symbols denoting nucleic acid products (genes, alleles, mRNAs etc) are italicized in print (or underlined in manuscript); symbols denoting amino acid products (proteins)should be represented in standard fonts. Italics need not be used in gene catalogs. To distinguish between mRNA, genomic DNA and cDNA the relevant prefix should be written in parentheses (mRNA)RBP1, (gDNA)RBP1, (cDNA)RBP1.
These symbols were formerly obtained from The Genome Database (GDB) and were assigned automatically to arbitrary DNA fragments and loci. These symbols comprise five parts e.g. DXS9879E, described by the following guidelines
a) D for DNA
b) 0, 1, 2,...22, X, Y, XY for the chromosomal assignment, where XY is for segments homologous on the X and Y chromosomes, and 0 is for unknown chromosomal assignment.
c) S, Z or F indicating the complexity of the DNA segment detected by the probe; with S for a unique DNA segment, Z for repetitive DNA segments found at a single chromosome site and F for small undefined families of homologous sequences found on multiple chromosomes.
d) 1, 2, 3,..., a sequential number to give uniqueness to the above concatenated characters.
e) When the DNA segment was known to be an expressed sequence the suffix E may have been added to indicate this.
App 1.2 Alleles
Allele terminology is the responsibility of the HUGO Mutation Database Initiative (now the Human Genome Variation Society). However, alleles should still be limited to an optimum of three characters using only capital letters or Arabic numerals. The allele designation should be written on the same line as gene symbol separated by an asterisk e.g. PGM1*1, the allele is printed as *1. When only allele symbols are displayed an asterisk should precede them. Sequence variation nomenclature, in the form of mutations and polymorphisms, is described by Antonarakis et al.  and den Dunnen and Antonarakis . HLA alleles are assigned by the WHO Nomenclature Committee for Factors of the HLA System  via the IMGT/HLA database . Immunoglobulin and T cell receptor alleles are assigned by the IMGT Nomenclature Committee via the IMGT/LIGM database .
App 1.3 Splice variants and promotors
The HGNC has no authority over protein nomenclature; however, we are frequently asked how to designate splice variants so we suggest the following:
Proteins should be designated using the same symbol as the gene, printed in non-italicized letters. When referring to splice variants, the symbol can be followed by an underscore and the lower case letter "v" then a consecutive number to denote which variant is which e.g. G6PD_v1. For genes which use multiple promoters, the promoters can be designated in a similar way to splice variants but using the lower case letters "pr", e.g. symbol: G6PD_pr1, name: "glucose-6-phosphate dehydrogenase, promoter 1".
App 1.4 Species designations
When necessary to distinguish the species of origin for homologous genes with the same gene symbol, the letter-based code for different species already established by SWISS-PROT should be used, commonly used species are shown in App 2: Table 2. The code is for use in publications only and should not be incorporated as part of the gene symbol. The species designation is added as a prefix to the gene symbol in parentheses. For example HUMAN signifies Homo sapiens and MOUSE signifies Mus musculus e.g. human genes: (HUMAN)ABCA1; (HUMAN)SLC13A2; homologous mouse genes: (MOUSE)Abca1; (MOUSE)Slc13a2.
|Greek Symbol||Latin upper case conversion|
|fruit fly||DROME||Drosophila melanogaster|
|puffer fish||FUGRU||Fugu rubripes|
|puffer fish||TETNG||Tetraodon nigroviridis|
|Amino acid||Three-letter symbol||One-letter symbol|
|Gln + Glu||Glx||Z|
Many thanks to Stylianos Antonarakis, Ségolène Aymé, Amos Bairoch, Richard Cammack, Panos Deloukas, Ian Dunham, Cara Hunsberger, Ian Jackson, Peter Little, Donna Maglott, Gerard Manning, Dan Nebert, Mark Paalman, Chris Porter, Kim Pruitt, Yoshiyuki Sakaki, Stephen Scherer, Alan Scott, Tom Shows, Cindy Smith, Nigel Spurr, Keith Tipton and anyone else who contributed with feedback. The HGNC is supported by NIH contract N01-LM-9-3533 and by the UK Medical Research Council.
1. Shows TB, et al. International system for human gene nomenclature (1979) ISGN (1979). Cytogenet Cell Genet. 1979 25(1-4):96-116. PMID:295268
2. Shows TB, et al. Guidelines for human gene nomenclature. An international system for human gene nomenclature (ISGN, 1987). Cytogenet Cell Genet. 1987 46(1-4):11-28. PMID:3507270 PDF
3. McAlpine P. Genetic nomenclature guide. Human. Trends Genet. 1995 Mar (suppl.): 39-42. PMID:7660472
4. White JA, et al. Guidelines for human gene nomenclature (1997). HUGO Nomenclature Committee. Genomics 1997 Oct 15; 45(2): 468-471. PMID:9344684 PDF
5. Lander ES, et al. Initial sequencing and analysis of the human genome. Nature 2001 Feb 15;409(6822):860-921. PMID:11237011
6. Venter JC, et al. The sequence of the human genome. Science 2001 Feb 16;291(5507):1304-51. PMID:11181995
7. Wain HM, Lush M, Ducluzeau F and Povey S. Genew: The Human Nomenclature Database. Nucleic Acids Res. 2002 Jan 1;30(1):169-71. PMID:11752283 PDF
8. White J, et al.Networking nomenclature. Nat Genet. 1998 Mar;18(3):209. PMID:9500537 PDF
9. Marsh SGE, et al. Nomenclature for factors of the HLA System, 2010. Tissue Antigens 2010 Apr;75(4):291-455. PMID:20356336
10. Robinson J, Mistry K, McWilliam H, Lopez R, Parham P, Marsh SGE. The IMGT/HLA Database. Nucleic Acids Res. 2011 Jan;39(Database issue):D1171-6. PMID:21071412
11. Lefranc MP. IMGT, the international ImMunoGeneTics database. Nucleic Acids Res. 2001 Jan 1;29(1):207-9. PMID:11125093
12. Mackenzie PI, et al. The UDP glycosyltransferase gene superfamily: recommended nomenclature update based on evolutionary divergence. Pharmacogenetics. 1997 Aug;7(4):255-69. PMID:9295054
13. Nelson DR, et al. P450 superfamily: update on new sequences, gene mapping, accession numbers and nomenclature. Pharmacogenetics 1996 Feb;6(1):1-42. PMID:8845856
14. Antonarakis SE. Recommendations for a nomenclature system for human gene mutations. Nomenclature Working Group. Hum Mutat. 1998;11(1):1-3. PMID:9450896
15. den Dunnen JT and Antonarakis SE. Mutation nomenclature extensions and suggestions to describe complex mutations: A discussion. Hum Mutat. 2000;15(1):7-12. PMID:10612815
Flowcharts that describe how the HGNC names protein coding, pseudogenes and ncRNA genes can be found here.