Gene symbol precision

Several gene databases, including heavily used ones such as the National Center for Biotechnology Information (NCBI) database, erroneously assign, on occasion, literature references to genes or proteins. These mistakes are mostly due to an overlap in gene aliases, whereby two distinct genes share a pseudonym. This is particularly confusing when the gene products have also biological properties in common, are part of signaling pathways that cross-talk to one another, or are regulated by the same effectors. We present examples spanning several research fields including apoptosis, ubiquitin-dependent degradation, signaling by Notch, Wnt, and small G proteins, transporters of glutathione conjugates of electrophiles, and mitochondrial and ribosomal RNA genes. To solve the problem, we argue in favor of including Entrez gene numbers in papers submitted for publication as unique gene identifiers to allow precise identification of genes and species studied.

► Unrelated genes can share aliases leading to homonymy and gene confusion.
► Erroneously assigned literature in databases can further add to gene confusion.
► Confusion very likely for homonymic genes operating in the same pathway or ontology.
► Attempts at unifying gene nomenclature have so far failed.
► A simple and pragmatic solution is proposed to eliminate gene confusion.