When a picture is worth a thousand base pairs …

Having introduced the subject of molecular barcodes as a means of identifying plants some months ago (“Berlin and barcodes”), I thought I should show you what a barcode actually looks like:


What we are looking at here is one small part of the gene that codes for the larger of the two subunits of RuBisCO, RbcL.   A, G, T and C refer to the four nucleosides (adenosine, guanine, thymine and cytosine) which make up DNA.   Think of this as the blueprint for one component of a type of diatom (Navicula lanceolata, in this case).   The carburettor in your car is basically the same as the carburettor in every other car, but a mechanic still has to make sure that the carburettor he fits to your car is the one that was designed for that model.  In the same way, the RuBisCo “component” differs very slightly between species.  Navicula_lanceolata

Six of the 45 populations of Navicula lanceolata collected and cultured for our DNA barcoding project.   All 45 populations differ by no more than three of the 1446 base pairs in the RbcL barcode.  The scale bar is 10 micrometres (= 1/100th of a millimetre) long.

If you have spent years learning the craft of identifying any group of organisms by traditional means, the use of molecular barcodes can seem like cheating.   The string of nucleosides reduces all the complexities of form and function to a charmless abstraction.   More importantly, you can’t see how the organism lives and how it interacts with the other organisms around it.   It epitomises all that is bad with reductionism in ecology.

Yet those of us who study diatoms are already offenders in this regard.  The focus of our attention for over a hundred years has been the silica “shell” (or “frustule”, to use the correct term).  The live organisms have to be “cleaned” with acids and oxidising agents to remove all the soft materials.   What we are left with is, itself, no more than a cipher of the true organism.  We look down our microscopes and a series of visual stimuli – length, breadth, outline, the presence of different types of mark upon the frustule – all link to particular names.   Diatomists have been using the empty shells, in effect, as “visual barcodes” for a long time.   Indeed, many diatoms have only ever been described in this state.

One of my long-term gripes is that diatomists have lost touch with the world of functional ecology. We name diatoms, count the number of each species, and then look for associations between species and particular features of their environment.   Navicula lanceolata, the subject of this post, has distinct habitats and preferences that can only be discerned from looking at the live organism (see “Coxhoe” from March 2013).    As a relatively large and distinctive diatom, it is quite easy to find under the microscope, particularly if you are willing to brave the cold conditions that it prefers.   Other diatoms are not so easy to recognise in their live state.  What nuances of their ecology are we missing as a result?