Sir

Most journals, including Nature, require authors to annotate a new entity (a gene, protein or loci, for example) with references to a standard database. However, journals do not require references to standard databases for discoveries of functions or diseases associated with previously defined genes. Since most genes have more than one name, and many gene names refer to more than one gene, the choice of a name without reference to a common or standard database can inhibit the integration of results from transcriptomics, population studies or comparative genomics.

In this post-genomic era, researchers have to be able to make associations among many genes, which requires being able to correctly identify a gene and all its synonyms. The most obvious way to ensure this would be for journals to insist that genes in a publication should be identified with reference to the Human RefSeq (see http://www.ncbi.nlm.nih.gov/RefSeq). In this way, genomic analyses are more likely to identify genes of common interest.