Gerstein et al. 2007

Gerstein MB, Bruce C, Rozowksy JS, Zheng D, Du J, Korbel JO, Emanuelsson O, Zhang ZD, Weissman S, Snyder M. 2007. What is a gene, post-ENCODO? History and updated definition. Genome Research 17: 669-681.

These authors reacted to some of the findings of the ongoing ENCODE project by redefining “gene”. Their new definition of this sometimes-contentious word is “"A gene is a union of genomic sequences encoding a coeherent set of potentially overlapping functional products."

They explain in the text first, a brief history of changing definitions of “gene”, from the coining of the term around 1900 to just prior to the publication of the ENCODE consortium (The ENCODE Project Consortium, 2007). Second, they describe complications and phenomena discovered by ENCODE and other research that render the current definition problematic. Third, they describe the important criteria in determining a new definition. These criteria are described as “backwards compatible”, organism-independent, statement of a simple idea, practical, and compatible with other biological nomenclature. The new definition meets these criteria.

The new definition also raises a more difficult question about function, and defining function in a biochemical and molecular context. The hard part is “what does this gene do?”, which can be answered (or not) at multiple levels.

Much of this paper reads like a corrolary to a strongly (though not strictly) adaptationist view of evolution, in addition to its strong (though not exclusive) human focus. One particular minor annoyance is a description of Ohno’s 1972 work, coining the term “junk DNA”, as intrinsically dismissive of all function of non-coding elements; this is not actually what Ohno (1972) said. Other adaptationist-leaning-statements include reference to the potential for many (most?) unannotated transcripts to represent transcriptional “noise” only in passing, seemingly as an afterthought, though Tress et al. (2007) are cited; I have not yet read that paper.

Overall, this new, ENCODE-informed definition seems useful in most contexts that I am likely to encounter the term “gene”.