1
$\begingroup$

I formally come from a physics/physical chemistry background, but I have begun to specialize in the area of the biomedical sciences and biomedical engineering. I found myself reading an article which talks about using leukocytes as carriers for targeted cancer drug delivery. One of the frequent adhesion molecules that comes up in the paper is E-selectin, whose wikipedia page I looked up to learn more about the molecule.

On that wikipedia page, it says that E-selectin is encoded by the SELE gene. Now, I can see that the gene is called "SELE" as it abbreviates "Selectin E", but my question is: how did this become the "official" name for the gene? Why doesn't the official name reveal anything about the actual nucleotide sequence of the gene? If someone had never heard of E-selectin before, how is he/she supposed to learn anything about the gene just from the name "SELE" ?

Sorry if this is a very nit-picky question. It's just that most of the substances talked about in chemistry have official names that reveal information about the structure, even if there is a more commonly used "nickname."

$\endgroup$
  • $\begingroup$ Biology names, especially proteins have been a free for all. The investigator that discovered it would usually name it, but there wasn't really any standard. Some named them by molecular weight, such as p21 or p53 which were 21 and 53 kD respectively. Even worse is that people studying different organisms would discover a protein and name it and it was only much later on that it was realized that the molecule had homologs in other organisms with different names. That is how you can get the Son of Sevenless or Sonic hedgehog genes or more descriptive names like DNA ligase or Katanin. $\endgroup$ – AMR Jan 16 '16 at 1:47
  • $\begingroup$ Often times we knew the proteins long before we determined the genetic sequence. With alternate splicing, you can end up with different proteins with different functions originating from the same gene. We estimate there are over 100,000 different human proteins, but only 20 to 30 thousand discrete genes. That means that some of this proteins are fusions of multiple gene products are the result of alternately spliced DNA. So the gene may have been named for a different protein and it was only later discovered that a different protein was a different splice form of the same gene. $\endgroup$ – AMR Jan 16 '16 at 1:50
2
$\begingroup$

Welcome to Biology.SE.

To my knowledge there is no good rule on how to name genes. The first person to name a gene or an allele is free to just pick up a that (s)he finds easy to remember and eventually intuitive. Sometime the author just picks a funny name. Here are a few examples from Drosophila melanogaster data base:

  • bag of marbles: involved in oogenesis, spermatogenesis.
  • breathless: involved in development of trachea.
  • buttonhead: regulates segmentation of the head.
  • dissatisfaction: involved in many aspects of sexual behavior.
  • hobo: a mobile genetic element.
  • Indy: for "I'm not dead yet"; this mutation makes the fly live twice as long.
  • windbeutel: German for "cream puff".

Did you suggest to name genes based on their sequence? Would be quite painful! One cannot name a gene based on its sequence as the sequence is way too long. The median gene length in humans is 24,000 base pairs (Fuchs G et al. 2014). You just need to refer to a database to match a sequence to a gene name (see BLAST for example).

How could we improve this system?

Do you have any suggestion on how we should name genes? Should they be named in relation to the phenotype they affect or to the biological pathway that are involved in or maybe something else? The issue is that we often annotate and name a gene before knowing much about its effect.

Note that some genes have several names which is indeed confusing. Would be good to solve this a little bit.

$\endgroup$
  • $\begingroup$ Thanks for the detailed answer! I wasn't suggesting that each gene by identified by its sequence because I know the lengths that we're talking about here. I was just hoping that a kind of "language" would be developed, where sequence could be conveyed in a few letters. I can't think of any good ideas off the top of my head, but let's look at a chemistry analogy: take for instance diethyl ketone. It would be cumbersome to specify an origin position and then provide coordinates for each atom. However, those 13 letters are enough to tell you it's (C2H5)2C=O. That's what I was talking about. $\endgroup$ – Muhammad Khan Jan 16 '16 at 2:18
  • 1
    $\begingroup$ To add to this... there is a transposon in Drosophila called Baggins and a gene called Smaug.. Both these are characters from the novel -The Hobbit $\endgroup$ – WYSIWYG Jan 16 '16 at 6:16
  • $\begingroup$ @MuhammadKhan Biological macromolecules are hundreds to thousands of amino acids long Genes can be a few hundred to well over a million nucleotides long, and they are interrupted, meaning that there are long stretches, introns, that get spliced out of the mRNA that makes the protein. You are asking more of an information theory question in How can we compress the sequence into a form that is manageable but can be decompressed into the full sequence, and that would be difficult. Just take a look at the protein FASTA sequence for SELE and you will get an idea of the challenge. $\endgroup$ – AMR Jan 16 '16 at 18:17
  • 1
    $\begingroup$ This is an example of a rather simple biomolecule, Acetyl Coenzyme-A, and its IUPAC name. S-[2-[3-[[(2R)-4-[[[(2R,3S,4R,5R)-5-(6-aminopurin-9-yl)-4-hydroxy-3-phosphonooxyoxolan-2-yl]methoxy-hydroxyphosphoryl]oxy-hydroxyphosphoryl]oxy-2-hydroxy-3,3-dimethylbutanoyl]amino]propanoylamino]ethyl] ethanethioate You can start to see the problems that biologists would face if they tried to adopt chemical nomenclature. $\endgroup$ – AMR Jan 16 '16 at 18:23
  • $\begingroup$ @AMR You're right in saying that the question boils down to something more of information theory rather than biology. I guess I'm just surprised that biologists and information theorists (or people at the intersection of these two) haven't made efforts toward this yet. For your chemical example, I know that names get cumbersome which is why names like acetyl co-A were invented, but at least there is a formal name one could analyze if need be. Gene names don't have "standard" names $\endgroup$ – Muhammad Khan Jan 16 '16 at 19:08
1
$\begingroup$

Unfortunately there are no rigorous naming conventions for a gene. While some are based on actual molecular function or phenotypic effect, others can be totally bizarre names. Some are actually based on the test-tube and purification fraction labels. Some are named on characters from Mythology, novels etc.

I'll add some examples:

SNARE (proteins involved in vesicle fusion): Soluble NSF Activated protein REceptor (where NSF is N-maleimide sensitive factor).

Nanog: Inspired from an Irish folk tale. There is a magical land called Tír-na-nÓg where if one goes, becomes immortal and forever young. So the discoverers of this gene which responsible for stemness of a stem cell, named it as Nanog.

In fact, a lot of Drosophila genes are named after the visible phenotypic effect of the gene mutation. For e.g. Hairy, Wingless, Sex Combs Reduced etc.

Many genes were actually discovered before the genomes got fully sequenced and since these genes were well studied, the idea of renaming them would not have been very appealing. However, yeast genes have been given formal systematic names.

From SGD

YAL001C first ORF to the left of the centromere on chromosome I (A is the 1st letter of the English alphabet), on the complement or Crick strand

YGR116W 116th ORF right of the centromere on chromosome VII (G is the 7th letter of the English alphabet), on the Watson strand

While I agree that absence of any naming convention would not convey any information about the function of the gene, these wild names do make studying biology interesting. Yeast systematic genes names are so uninteresting and hard to remember (at least for me).

However, there are standard nucleotide repositories such as NCBI, ENA and DDBJ which maintain records about different genes. Moreover, there are standard organism specific databases such as FlyBase for Drosophila, TARE for Arabidopsis etc. In these databases you will find the standard names and description of these genes.


Also have a look at this post: Are there more descriptive ways of naming genes and gene interactions?

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.