Thursday, March 21, 2013

Walk on the wild-type side

Let's say I want to test the effect of gene A that I've modified somehow -- to mimic a human disease, say -- on a mouse (assuming, of course, that the results are then generalizable to people).  I put the modified gene into the zygotes of a laboratory mouse strain, let's say C57/Bl, and watch what happens as the animals develop.  Of course I need controls, unmodified mice, so that I can compare my transgenic modifications to the unmodified C57/Bl.  When I write my experiment up, I will refer to the 'mutant' (transgenic) mouse and the 'wild-type'.

Wild-type axolotl (Wikimedia)
Indeed, everybody everywhere in biology, it seems, refers to their manipulated or in some way conditional organism as the 'mutant' and the other, non-manipulated form as the 'wild type'.  This is so entrenched that often even seasoned professionals don't know, don't think about, or don't want to be forthcoming about what that means.

As judges, we just heard a student research presentation in a high-school science project competition that we wrote about on Tuesday, in which the student compared the effect of a modified protein to that of the wild-type.  This student was plunked into a fancy lab and clearly got this term from her mentors in the lab, without knowing where it came from (other than that the protein "was cloned into E. coli") or thinking about what that meant.  We would assert with nearly 100% confidence that the mentor never thought about it either.

But, who cares?  
Scientists like other scholars or, perhaps, those in any profession, coin jargon to use in their work.  Sometimes a term is clear and stands for a specific thing, but says it more simply.  We can say, for example, that the frequency of a genetic variant in a population 'drifted' over time.  What we clearly mean is that "the allele frequency changed randomly over many generations because it did not cause an effect on the organism that carried it that affected that organism's survival or reproduction".  Just saying 'drifted' saves lots of words.  Properly used, such technical terminology is completely reasonable and useful.

Sometimes we are using terms to show off, to seem intelligent, insightful or perceptive, or to give an aura of technical expertise.  The humanities--we're sorry to have to assert--typically wallow deeply in terms so contorted and arcane that even they don't seem to agree on their meanings. That is, the jargon is obscurative rather than efficient or clarifying.
Mutant axolotl (reduced pigmentation); Wikimedia

But sometimes even in natural science, terminology becomes an obstacle rather than a shortcut.  Many analysts of science and the philosophy and history of science have noted how deeply we can be embedded in rhetoric which is not just shorthand, and not just showing off, but which carries connotations that actively affect how we think and what we do.  Technical terms can allow airy underpinnings to be taken as having substance.  Then it becomes quite dangerous to science (even if, perhaps, it helps get grants and your name in the paper)!

In our areas of genetics and evolutionary biology, as a rule, terms at least have somewhat of a precise meaning, even if referring to, say, the 'wild type' sounds like insider talk.  This particular term might seem to be absolutely no problem, since it's so simple.  But what is the 'wild type', even of a single gene?  Is it 'the' sequence of the gene found in the wild?  Almost never!  Why?  Because the 'same' gene will have varying sequences among individuals in a population, or in a species, or, even more so among species.

So to say you're comparing the risk of a variant to 'the' wild type is to make an assumption that there is such a type.  Do you mean the 'normal' variant?  If so, how do you define normal, and by what reason do you assume, much less actually think, that there is only one such type out there in the 'wild'?

Type specimen 'pitcher plant' (Wikimedia)
The term originated, we think, in the early 20th century experimental biology, when evolutionary theory had it that most genes had a good variant that was by far the most common in the population, and one or a very few, very rare, harmful 'mutant' forms.  Nature preferred the former, so it was the 'wild type,' and in a lab one would naturally use that form (and term) to refer to what originally was sampled from the wild, and so on.

However, we know that was a very poor understanding of evolution, of the nature of DNA sequence variation, and it was a hyper-Darwinian stereotypical treatment of nature.  It was, we think, entirely analogous to the idea of the 'type specimen'.  That's the bedraggled stuffed lion in the Natural History museum, taken to represent all lions.  Of course, if we're careful in our thinking, we know that no other lion is quite like the Snagglepuss in the museum.  But science is not always careful--in part because the rat-race for status in the field encourages simpler, if not simplistic thinking, in which we scientists routinely indulge.

If you're studying the mice in your laboratory, comparing the animals of the strain you tinkered with by introducing an altered gene to those in the strain that you did nothing to, then you're likely to call the latter the 'wild type'.  But that is preposterous!  There is nothing 'wild' or 'typical' about a laboratory mouse!  They are the result of, at the very least, many generations of inbreeding and cage maintenance and as the old quip goes, wouldn't survive for more than 5 minutes out in the, yes, wild!

Walk away from the wild-type side of thinking!
If we really want to understand Nature, we should wean ourselves from habits that can be very, if subtly, misleading. We've often mentioned the related and comparably misleading ideas of genes or selection 'for' a given trait, or the sloppy and slippery assumption that every trait was produced by natural selection, or naming genes 'for' some trait (like 'bithorax' or 'presenillen' for genes that, when mutated in particular ways in particular individuals can cause, respectively, a double thorax in insects, or early onset Alzheimer's disease in humans).  Or our routine reference to 'the' human genome--a sequence that we've posted about before, that doesn't exist in a particular individual, never has, and never will.

We often have no need whatever for special terms, and we should stay away from them like the plague.  If we want a term, then instead of the stereotypical notion of a 'wild type', that can end up not being properly checked or used in interpreting results, we should use terms like 'reference', or 'baseline',  or 'control', or even, if we're careful, 'average'.  Those terms clearly state what you mean, without latent connotations that can affect your audience's thinking....or even your own.  You can choose an inapt referent, or the average may or may not be a good baseline, but at least the meaning should be clear and your conclusions then fairly discussed.

So to keep from being wildly misled, don't walk on the wild-type side!


Anonymous said...

Hi, 'wild type' is actually the subject of my historically-minded philosophy thesis. My research is ongoing, but I can tell you that the idea of a wild type is pretty old - my earliest reference is the 1820s. However, it's only with the development of more hereditarian perspectives which don't consider type as intimately entangled with environment that you get the idea that wild types can be taken out of nature and brought into the lab. Before then, wild type generally referred to the wild ancestors of domestic varieties. Of course, some geneticists have always been aware of the fact that lab conditions and selective practices considerably alter the organisms they use from their original wild nature. As to why they accepted the use of the term 'wild type', I think in TH Morgan's case you can see a certain instrumentalism in his defence of his lab's techniques in his 1916 'Critique of the theory of Evolution'. Basically, he says you can only get secrets out of nature if you torture them out of her. That lab conditions might not be representative of natural ones and lab 'wild types' very different from real wild members of the species doesn't mean, in Morgan's view, that useful results can't be produced, in fact this is the only way useful results can be produced given the limitations of more naturalistic approaches. As long as the lab wild type can 'stand in for' 'normal' wild individuals, its use isn't problematic. Of course, these views were aired when genetics had a much more limited repertoire of techniques and methods available. Naive treatment of the wild type as representative of the species norm may therefore today be regarded as less forgivable and in fact an obstruction to the carrying out of good science, whereas in Morgan's day such abstractions and idealisations may be considered as having been somewhat necessary.

Ken Weiss said...

Thanks for this interesting comment. Wallace's paper almost had wild-type in its title, and I know that 'type' was an old term. Type and variety were, I think (you would know better than I) that were in the working repertoire of agronomists and naturalists, probably from Linnaeus if not even much earlier (Aristotle? Platonic ideals?).

I associated the term 'wild type' in my own usage here, with first the idea that you bring it into the lab, as you say, and then you tinker or compare it to something that's been manipulated (or whose living conditions or whatever have been). But second, with whatever you haven't yet manipulated.

I don't know the details of Morgan's era the way you do, but the idea of type specimens was antecedent and had to be related to the 'wild type' phrase as a way to compare lab work to the museum's specimen.

I think that, like so many of us in so many ways, we may 'know' various distinctions and subtleties, but that doesn't mean we don't lapse readily into comforting metaphoric simplification. I would also assert, but of course can't prove, that many--many!--in genetics and other lab work do not really understand the important points.

I can't do more than just suggest that this has to be somewhat true when people refer to, for example, a C57 mouse as the wild-type.

Interesting, the points you make about control may be turning full circle in many ways. A prominent geneticist just told me in discussion last week that he thought experimental breeding was so problematic that natural populations--humans in this case--were the proper material to study.

I'll be interested to see what you say in your thesis when, presumably, it's published in one form or another.

Ian Dworkin said...

This was a great post, and useful comments. I am curious if either of you could provide some of the first references to use the term "wild-type"?


Ken Weiss said...

No, I can't because I've not tried to track it back. One of the above comments suggests early usage of similar term. I would check a few history of biology books (I have a few at home), but they may not even have considered this anything to write about (so ingrained is the thinking).

But there was an article in Science by C. Schuchert in 1897 ('what is a type in natural history') and a reply later in the year in the May 7 issue, a letter called 'type specimens' in natural history.

But I found by Googling that this may go back to early Mendelians (see Wild-type First Draft II, to get a file of a paper that discusses this).

That's the best I can do!

Anne Buchanan said...

I'm wondering if Thomas Morgan was the first to use it. I'm trying to run that down.

Ken Weiss said...

Whatever we (or Ian) can track down, the idea of a 'type' for a species certainly goes back at least to Linnaeus. I would imagine it clearly can be found in Aristotle, and it was quite central to Plato's epistemology.

So in a sense it is the addition of 'wild' as a contrast to modified-in-the-lab that seems to be the question. And then that was extended to mean unmodified (by the investigator) vis-a-vis modified (by the investigator) that we think is quite misleading. C57 or DBA etc. mice are not real mice!

Tarquin Holmes said...

Dear Ken,

It's been a while since I made my original comment (3 years!) but I can finally get back to you to say that I have completed my thesis and that it is now openly available through my university's open research archive:

To reply to Ian's query regarding early references, see chapter 4.

If anyone reading this blog would be interested in a more concise summary of the history of wild type, I have a draft of a paper I am working on for journal publication which summarises large sections of the thesis. i cna be contacted at