We all know what a genome is, and we think we understand the term proteome, but can anyone tell us the constituents of a functome? As the availability of complete genome sequences has spawned analyses of entire complements of proteins, RNAs, metabolites, and other cellular constituents, there has arisen a need for a terminology expansive enough to encompass the global scale of the data. A sensible suffix was appropriated for this purpose, but now is proliferating uncontrollably: genome, proteome, transcriptome, metabolome, interactome, even phenome, with many more 'omes sure to be in various stages of gestation. Perhaps it is not completely coincidental that 'ome is also the anglicized form of 'oma (1), commonly used to name such unwelcome intrusions as sarcoma, lipoma, and fibroma. This metastatic growth of the 'ome is spreading imprecision and confusion. Meanwhile, research summaries in the front of major scientific weeklies with titles like 'Ome Sweet 'Ome, and 'Ome...'Ome...'Ome: The Genomicist's New Mantra, and our personal favorite, The 'Ome: A Pièce de Résistance, only serve to confuse us further. Because a clear and widely accepted nomenclature is essential for the health of any discipline, a systematic solution to this problem is urgently needed.
It is often instructive to look to the past for guidance. A now familiar nomenclature grew up around the related suffix 'some (for which 'ome is sometimes mistaken), meaning "body," which has been used to name various intracellular particles. "Chromosome" dates back more than 100 years, "ribosome" and "lysosome" are nearly half a century old, and "nucleosome" and "replisome" originated more than a quarter century ago (1). Even "spliceosome" and "proteasome" are approaching two decades of service. This relatively modest growth in the application of the suffix 'some contrasts with that for 'ome. Although "genome" was coined by German scientists ("genom") in 1920 and first used in English in 1930 (1), none of the other 'omes can lay claim to more than a few years of history.
There are two underappreciated and so far unresolved predicaments with the 'ome terminology. First, there is a problem with its scope. Whereas the extent of the genome is clear (all the genetic material of a cell), what constitutes a transcriptome is not so obvious. Is it just the mRNAs, or does it include the transcripts produced by RNA polymerases I and III? What about transcripts that end up in the enzymes telomerase or RNaseP, or in ribonucleoprotein particles such as snRNPs? The precise constituents of an 'ome are often not well specified.
A second, and much more severe, problem is the conditional nature of some 'omes. The genome--notwithstanding the occasional hop of a transposon or rearrangement of an immunoglobulin gene--is a relatively fixed entity, and reasonable people can agree on its definition. But the proteome present in a cell at one moment will differ drastically from that in the same cell moments after it has been heated to 65ºC. Or, if we define a cell's glycosylome at time zero, and seconds later the cell undergoes programmed cell death, its carbohydrate moieties are likely to give up the ghost nonuniformly, with some persisting to the last. At what point in this process do we define the glycosylome?
To circumvent these difficulties and others sure to emerge, we propose some simple rules. First, considerable precision can be gained by a more circumscribed representation of the 'ome's constituents, for example, phospholipidome rather than lipidome; inositol phospholipidome rather than phospholipidome. Of course, this has the potential to be abused and to lead to absurdly finer subdivisions. For example, do we want the transcription factorome to be subdivided into the transcriptional activatorome and the transcriptional repressorome? Does not the transcriptional activatorome then include the zincfingerome, which itself includes the Cys-His zincfingerome and the Cys-Cys zincfingerome? To avoid this pitfall, we propose that the minimum number of similar cellular constituents that constitute an 'ome be clearly defined. Seven or eight seems to us a conservative yet valuable cutoff. Thus, there can be no "nucleicacidome" (there's only DNA and RNA, after all), but there certainly is a "nucleotideome" (A, T, G, C, U, I, plus myriad modified purines and pyrimidines); no "actinomae," of course (humans have only six actin isoforms), but definitely a "tubulome" (multiple a and b tubulin isotypes, not to mention g, d, e, z, and h tubulins).
Second, it would be helpful if the state of the cells for which an 'ome is defined were apparent in the nomenclature. If initially we use basic parameters like temperature, pH, cell cycle stage, and subcellular localization, we can obtain a definition such as, "the 37º-7.4-G1-Golgi-N-but-not-O-linked glycosylome." This system has enormous versatility and can be suitably expanded to incorporate other parameters, including cell source, developmental timing, and much more. Perhaps at first this may seem a bit cumbersome. But please remember that this nomenclature is no more intricate than the (±)-N-methyl-g-[4-(trifluoromethyl)phenoxy]-benzenepropanamine used so effectively by chemists and many others. As a more wieldy alternative, we also propose that an Enzyme Commission (E.C.)-style nomenclature should be established that allows the incorporation of as many specifications as needed. In this format, the particular glycosylome mentioned above has been provisionally designated the 4.7.5.3.8ome. We expect that these numerical names, after a sufficient number of citations, will become as familiar as many E.C. numbers.
The adoption of our simplified system means that as new technologies emerge enabling the assay of yet more cellular constituents, the nomenclature is already in hand to deal with the discoveries. Which brings us to a final thought. As biologists approach a definition of all of the various machines that carry out life's basic processes, we should be able to define the ultimate 'ome, the collection of all of these machines: the "someome." Others might prefer to call this the "omesome," given that it defines the machine comprising all the 'omes. Either one is a vast improvement over their imprecise and prosaic synonym currently in wide use: the cell.
Reference
S. Fields is in the Department of Genome Sciences and Department of Medicine, Howard Hughes Medical Institute, University of Washington, Seattle, WA 98195, USA. M. Johnston is in the Department of Genetics, Washington University Medical School, St. Louis, MO 63110, USA. E-mail: fields@u.washington.edu, mj@genetics.wustl.edu
|
If you have come to this page from an outside location click here to get back to mindfully.org |