Of alleles, loci and gene nomenclature

Weeden, N.F.

Department of Horticultural Sciences
Cornell University, Geneva NY 14456, USA

The principal reason for developing a nomenclatural system is to enhance communication. Hence, it is extremely important that we agree on what common terms and symbols mean. Recently, I was asked to give a short presentation of gene nomenclature in pea (Mike Ambrose probably should have given the presentation, but he could not make the meeting). Before giving the presentation, I consulted with Mike as well as several members of the Coordinating Committee who had expressed an interest in changing certain practices (e.g. see Murfet [2]). A number of minor changes has been implemented within the last year, in particular the elimination of hyphens within a locus symbol (e.g. Adh1, not Adh-1 and Aatp, not Aat-p), and these have been incorporated into the current volume of Pisum Genetics. However, other changes have been suggested, and I would like to present here a few concerns that came to mind as I was reviewing the situation in pea and listening to the situation in other species.

A problem that surfaced immediately was with terminology. I think we can all agree on a definition for ‘allele’ and ‘locus,’ but I have my doubts about ‘gene’. In A Dictionary of Biology [1] ‘locus’ is defined as “a position in a chromosome.” ‘Allele’ is defined more indirectly: “Two or more genes are said to be alleles if they occupy the same relative position (locus) on homologous chromosomes...”. I would prefer something a bit more specific such as ‘the DNA sequence at a locus.’ However, in my reading of the literature and in my discussions with other geneticists I have found little disagreement about these relatively precise and non-overlapping definitions: locus refers to a geographical, sequence-independent position and allele refers to a specific DNA sequence at that position.

Such is not the case with the term ‘gene’. In the dictionary mentioned above it is nearly synonymous with allele (the actual definition is lengthy, but refers to a ‘piece of a chromosome.’ In contrast, I was told quite bluntly by a respected geneticist that 99% of all geneticists define ‘gene’ as . . . and proceeded to give me a definition that was similar to but not identical to the definition for locus given above. In the literature, the definition of gene appears to depend on the author. Speiss, in his book Genes in Populations [3], clearly uses ‘gene’ as a synonym for allele (gene frequency, genotype, etc.). Other authors uses the terms ‘gene’ and ‘locus’ interchangeably. Many authors, including several submitting to Pisum Genetics, use the term ‘gene’ in a more general sense, referring to the locus at certain times and to an allele at others. For instance, a researcher will describe a locus and immediately suggest “cloning or sequencing this gene.” In this case ‘gene’ is used both to refer back to the locus and to an undefined allele, the locus being only a position on a chromosome (or map), lacking a sequence and incapable of being cloned.

My purpose here is not to take issue with the usage of the term ‘gene’, which I think has become far too entrenched in genetic literature to change now. Rather, as Editor of Pisum Genetics, I want to point out the ambiguity of its meaning and to encourage those submitting articles to use the more specific terms ‘allele’ and ‘locus’ when possible. In these pages ‘locus’ will be used only to designate a position, and I will attempt to restrict the usage of ‘gene’ to something approaching synonymy with allele.

The second point I would like to discuss is based on the above differentiation of locus and allele. Traditionally, locus symbols for mutations have been either capitalized or left as lower case depending on whether the mutant allele is dominant or recessive, respectively. This tradition has been adhered to in the consensus map. However, the system was not closely followed in the submissions I received for Vol. 30 of Pisum Genetics. In some cases, loci were submitted in all lower case, in some, all capitalized, and some followed the traditional mixture. If the membership doesn’t really care, or if a change is desired, I would like to propose the following:

Given (1) that a locus is a position and not a sequence, then the dominant/recessive relationships of alleles at that locus need not and probably should not be represented in the symbol for the locus, and (2) that loci with co-dominant alleles or alleles that have been initially identified by protein or DNA sequence polymorphisms are generally given capitalized symbols if the locus is on nuclear DNA and lower case symbols if the locus is on an organellar genome, I suggest that all loci on the nuclear genome in pea be designated by capitalized symbols. Capital and lower case symbols could still be used when referring to dominant and recessive alleles at a locus; however, as DNA sequencing becomes more prevalent, we will find ourselves with more and more allelic series in which dominance may become ambiguous or at least dependent on the alleles being compared. The article by W. K. Swiecicki on pp. 22-23 of this volume was submitted in this format and has been left that way for review by the readership.


1. Abercrombie, M., Hickman C.J. and Johnson, M.L. 1951. A Dictionary of Biology. Penguin Books, Baltimore.
2. Murfet, I.C. 1997. Pisum Genetics p. 47.
3. Speiss, E. B. 1977. Genes in Populations. John Wiley & Sons, New York.