Lost in translation: On naming babies and genes

21 February 2014 by Malcolm Campbell, posted in Biology, Science, Science Communication

Our classifications are often plainly influenced by chains of affinities.” From On the origin of species by means of natural selection, or the preservation of favoured races in the struggle for life (1859) by Charles R. Darwin (1809-1882)

Sometimes words get in the way.

Consider the naming of our youngest son.

His name is Connor.

This would not be a problem in and of itself, were it not for the fact that he was named in the country of his birth, France. connor1That, and poorly executed attempts at pronouncing his name with a French accent by his father.

When Connor was born, we were informed that he had to be registered officially with the local authorities, the mairie. Not wanting to fall afoul of the authorities, a trip to the mairie was made with due haste.

At the mairie, the office for registering newborns lay at the end of a cavernous, dimly-lit corridor. Bright light slanted out of the office into the corridor, in a foreboding fashion. A small queue of men, presumably recently endowed with a new baby, lurked as silhouettes in the angular  light. Taking a position at the end of the queue, there was a short wait as each man stepped forward to register their child.

Two desks were arranged side by side in the office. Each desk faced the doorway, and behind each desk sat a prim, middle-aged woman with a ledger. In front of each desk was a chair for the registrant to sit and answer questions.

“What is the name of your child, monsieur?” Was the simple question asked, in polite French.

“Connor.” I stammered, with an attempt at a French accent.

The woman’s head tilted up from the ledger to look me in the eye.

“Pardon, monsieur?”

“Co--nnor.” A little more accented this time.

“Une autre fois, monsieur.” Another time, sir.

Clearly the name was causing some problems.

Perhaps an even more accentuated “French-sounding” pronunciation was in order.


The prim woman’s eyebrows shot upwards.

She turned to her colleague.

“Giselle, Giselle – écoutes, écoutes.”

Her colleague, Giselle, looked up from her ledger and turned to listen.

“Encore. Une autre fois, monsieur.”


Both women appeared aghast.

“Ce n’est pas possible, monsieur!” That’s not possible, sir.

An attempt to explain that, indeed, this was the name desired for the child, was met with furrowed brows.

A paper form was extracted from a drawer in the woman’s desk.

Politely, patiently, the woman explained that a declaration, an attestation, was needed. There needed to be proof that this name was acceptable.

“Acceptable, how?” I asked.

Acceptable in his parents’ home country, of course.

Clearly the name was not acceptable in France.

Carefully, “Connor” was written on the paper, and passed back across the desk.

After a focused read, a knowing realisation came to the woman’s face.

“AAAH! CONE-ORE! Comme Jimmy Connors!”

Connor, like Jimmy Connors, the tennis great. It wasn’t a cultural reference that had even figured in the naming of the baby, but whatever worked for the bureaucratic machinery was just fine.

The form was completed, signed, and approved. Jimmy Connors’ namesake safely registered. Grateful, but perplexed, a hasty retreat was made from the registration office.

Back at work with French colleagues, the response to Connor’s name played itself out again: “That can’t possibly be his name!” “Oh, you joker!”

Only when the name was written out on paper were things resolved. It emerged that an Anglophone-attempt-at-a-French-pronunciation was the root of the problem.  With the rather poor “Francophonisation”, “Connor” emerged as “connard” – French for “idiot” or “jerk” – as in “C'est un vrai connard!” – “He's a real jerk!”. No wonder Francophones had a problem with the name. Imagine someone naming their baby “Idiot”! Well, there was certainly one idiot outed on that day. connor2

The baby naming experience was truly an instance of something being lost in translation. A classic case of the real intention for the word – a Celtic name for our lovely auburn-haired baby – being completely different from that which is heard and understood – an idiot.

The communication of science is full of such pitfalls for miscommunication – instances where meaning can be lost in translation. As scientists, we make use of words that make great sense to us but whose meaning is lost or distorted when shared with people whose expertise lies outside of the realm of our specific disciplines. We have a specific understanding of the word – its origins and its meaning. It’s not fair to assume that others will infer this same meaning when they hear the word with their own frame of reference.

The naming of genes, like the naming of babies in another language, is a great example of where meaning can get lost in translation.

Gene names pose challenges for scientists and non-scientists alike. Consider the human genome, for example. The human genome has approximately 21000 genes. The shear number of possible gene names poses a problem in and of itself. How to give each gene a meaningful name that would convey the same information to experts and interested non-experts alike? Location in the genome might be a useful place to start.

Each gene has a particular location on a chromosome. Theoretically, each gene could simply be given its name on the basis of its location along the length of a given chromosome. This is one way of naming a gene – based on chromosome location – and, in fact, each gene in the human genome does have such a name. This said, it is better to consider this name more like an address than a proper name. Indeed, this name is considered the gene’s “map location” – the region where it sits on a map of the human genome. The reason why the location is not used as the name for the gene is that it is relatively uninformative with respect to the function of the gene and its functional relationship relative to other genes. That is, naming genes by map location alone would be the same as naming people on the basis of their street address – it says nothing about their relationship to each other.

One way to connect a genes name to its function would be to assign it on the basis of the product that is encoded by the gene – the gene product. For example, a large proportion of genes in the human genome encode proteins – macromolecules that serve specific functions within the cell. A significant portion of these proteins are known as enzymes. Enzymes function to catalyse a particular chemical reaction – converting one molecule to another, fusing molecules together, or cleaving them apart. Enzymes have a well-defined nomenclature, based on a well-established set of rules. Enzyme names are not impacted by the organism from which the enzyme was derived – so there is nothing special about the name of an enzyme called pyruvate dehydrogenase from Homo sapiens relative to the same enzyme from Escherichia coli. Given this, genes could have names designated on the basis of the enzymes they encode.

There is an effort to apply enzyme-naming conventions to genes. The Guidelines for Human Gene Nomenclature lay out a set of conventions to name genes according to the enzymes they encode. Consider, for example, the enzyme group known as cytochrome P450 monooxygenases. As their name implies, these enzymes use a co-factor, known as cytochrome P450, to catalyse reactions that add oxygen to organic molecules. Each cytochrome P450 monooxygenases is known as a CYP. CYPs are grouped into families contingent on the type of organic molecule it acts upon. Humans have 18 families of CYPs that are, in turn, divided into 43 subfamilies in total. All told, humans make 57 different CYP enzymes. Genes can be named on the basis of the CYP they encode. For example CYP19A1, a member of the CYP19 family, encodes a enzyme that catalyses the conversion of androgens to estrogens.

Naming genes on the basis of enzyme function seems sensible. However, this naming convention is limited. It is limited by the fact that only a subset of all genes encode enzymes. What’s more, it is limited by the fact that biochemical function is only one way to define a gene’s activity. While enzyme function is relatively unambiguous, it says nothing about the function of the gene in a larger biological context. It is the equivalent of giving each member of your family a surname and a number. It says nothing of the unique “personality” of the family members, and the broader role they play.

Not surprisingly, scientists have most frequently assigned gene names on the basis of biological function. Frequently, this function has been assigned by mutant analysis. As its name implies, mutant analysis involves exploring gene function by investigating the consequences of mutations in a given gene. Usually, mutant analysis begins by identifying individuals within a population that have an identifiable, inherited difference in a particular trait – that is, they are mutants. For example, mutants may be visibly different, they may have health defects, or they may be more susceptible to disease. These traits are inherited generation after generation. They have a genetic basis. The genetic basis is a consequence of a mutation in a particular gene. The mutation is a change in the gene’s DNA code that creates either a difference in the way the gene product functions or a change in the abundance of the gene product.

When the mutation that underlies a mutant trait is discovered, the gene in question is usually given a name based on the mutant. That is, the gene name relates to the effect that gene has when it isn’t functioning properly. It is for this reason that plants have a gene called LEAFY that is actually involved in making flowers. When LEAFY loses its function due to mutation, plants don’t make flowers, but leaves instead. In some ways, the name LEAFY says more about what the gene doesn’t do, than what it actually does.

Similarly, mammals have a gene called SONIC HEDGEHOG, which is decidedly not involved in turning mammals into video game characters. SONIC HEDGEHOG owes its name to the discovery of a fruit fly gene, HEDGEHOG, which is also definitely not involved in making fruit flies turn into hedgehogs. When HEDGEHOG is mutated, fruit fly embryos harbouring two copies of the mutation in their genome are not smooth. Instead mutant embryos are covered with small pointy protrusions, known as dentricles, giving them a hedgehog-like appearance (if you use your imagination).

When a mammalian equivalent to HEDGEHOG was discovered, it was dubbed SONIC HEDGEHOG.  Of course, the full name, SONIC HEDGEHOG, owes its etymology to the video game character “Sonic the Hedgehog”. connor3

The SONIC HEDGEHOG gene encodes a soluble protein that diffuses through the mammalian body. As it spreads within the body, the SONIC HEDGEHOG protein establishes a concentration gradient – high in some locations, and low in others. The SONIC HEDGEHOG protein gradient functions as a signal to direct the formation of animal tissues – making sure that the right things develop in the right places. The SONIC HEDGEHOG protein is, therefore, said to control animal morphology, and it is referred to as a morphogen.

It turns out that HEDGEHOG has relatively ancient origins, with functionally equivalent genes found in organisms as diverse as insects, fish and humans. As HEDGEHOG-related genes in different species emerged, so did some entertaining names. This included tiggywinkle hedgehog, named after Mrs. Tiggy-Winkle, from Beatrix Potter's children’s story, and echidna hedgehog, named for the monotreme, the spiny anteater. While they can still be found in the scientific literature, the general convention is that these latter two names should be pared back to variants of SONIC HEDGEHOG, and INDIAN HEDGEHOG – much more sober gene family names.

SONIC HEDGEHOG underscores the potential to lose something in translation with a gene name. While entertaining, and while it provides reference back to the original discovery in fruit flies, the name SONIC HEDGEHOG is largely irrelevant to its function in vertebrates such as humans, where it plays an important role in the development of the central nervous system. In humans, mutation of the SONIC HEDGEHOG gene can cause holoprosencephaly, a cephalic disorder where the forebrain of the embryo fails to develop into two hemispheres. Some have made the case that “entertaining” names like SONIC HEDGEHOG can undermine the seriousness with which the condition that it is involved in is taken. The gene name can be seen as belittling factor. What’s more, it can serve as a distraction, or even misleading, as to what the gene actually does.

Gene names are just one example where caution might be warranted in terms of better communicating science. On the one hand, catchy and entertaining names and terminology could, theoretically, enliven discourse around the science, and, potentially, serve as a useful means to remember some scientific information – an aide memoire. On the other hand, such words could obscure understanding. They come with their own history, their own meaning, and potentially their own baggage. They may appear pithy to us, or they may add another dimension to the work we undertake, but we must exert some caution lest we make pronouncements that, like the poorly-accented pronunciation of “Connor”, just sound idiotic.

Images: All photographs by Malcolm M. Campbell.


Echelard Y, Epstein DJ, St-Jacques B, Shen L, Mohler J, McMahon JA, McMahon AP (1993) Sonic hedgehog, a member of a family of putative signaling molecules, is implicated in the regulation of CNS polarity. Cell 75: 1417–30

Gray KA, Daugherty LC, Gordon SM, Seal RL, Wright MW, & Bruford EA (2013) Genenames. org: the HGNC resources in 2013. Nucleic Acids Research 41: 545-552

Marigo V, Roberts DJ, Lee SM, Tsukurov O, Levi T, Gastier JM, Epstein DJ, Gilbert DJ, Copeland NG, Seidman CE (1995) Cloning, expression, and chromosomal location of SHH and IHH: two human homologues of the Drosophila segment polarity gene hedgehog. Genomics 28: 44–51

Nüsslein-Volhard C, Wieschaus E (1980) Mutations affecting segment number and polarity in Drosophila. Nature 287: 795–801


connor & mom

Leave a Reply

9 + = eleven