Photo: Glasswing butterfly, probably Greta oto, on Asclepias curassavica; Eddy Van 3000 at flickr.com CC BY-SA 2.0
We could quit now, with our eyes on that glasswing butterfly: of course biology can be beautiful. Birds of paradise, lynx, ladyslipper orchids, Spanish moss*, orcas; can there be any doubt? But that’s not really what I mean. Is biology as a science beautiful, the way math is beautiful, and physics is beautiful?
Mathematicians talk about beauty a lot, and what they mean by beauty is a kind of simple elegance of structure in the way math works out. Take Euler’s identity: eπi+1=0. This expresses a simple relationship between the three most fundamental constants in math (e, π, and i) and the two natural numbers you need to build a number system (1 and 0) using the three basic arithmetical operations (addition, multiplication, and exponentiation) – and each appears exactly once. It’s stunningly beautiful, and the fact that the universe is structured so that it’s also true is astonishing. This happens enough that to many mathematicians the beauty of a theorem – or of its proof – suggests its correctness.
Physics is beautiful, too. The messy zoo of subatomic particles is explained by combinations of just six kinds of quarks. Energy and matter are forms of each other, and their interconversion is elegantly expressed as e=mc2. Planets orbit the sun in ellipses, and slow down and speed up such that a line from planet to sun sweeps out equal areas in equal times. There seems no particular reason our universe has to have the structure that makes these things true, but it does, and that’s beautiful.
But what about biology? On the face of it, biology seems far from beautiful. Ecologists wring their hands over the apparent lack of general laws – not just beautiful ones, but any general laws. Cell biologists deal with confusing webs of regulation, thickets of promoters and enhancers and repressors and activators and post-translational modification and phosphorylation. I could go on.
Biology is in many ways the science of kludges. Evolution has left it that way: time after time, faced with a problem, evolution has spurned elegant design in favour of weird and clumsy solutions. Stephen Jay Gould famously wrote about the panda’s thumb (pandas use opposable “thumbs”, but they aren’t the same thumbs we have; instead, they repurpose a bone called the radial sesamoid, and using different solutions to the same problem is surely inelegant). Humans get sore backs because our anatomy had to compromise from a design for walking on all fours. Whales, seals, and otters are aquatic, but remain tied to the surface by air-breathing lungs.
The list of kludges goes on and on, but perhaps the best example is the genetic code. The four bases of DNA (adenine, thymine, cytosine, and guanine), strung out like beads on a tremendously long necklace, code for the sequence of amino acids in constructed proteins. Of the many possible amino acids, life on Earth uses 20, so a code needs to use at least 3 DNA bases to represent each one (a 2-base code would give only 4×4=16 possibilities, which won’t do). In 1957, Francis Crick and colleagues hypothesized a coding system (nice explanation here) in which 3-base codes represented the 20 amino acids – with a 1:1 relation of code to amino acid and no ambiguity about where to put the “reading frame” dividing a string of bases into a string of 3-base codes**. It turns out that with these constraints, the very best 3-base code can represent exactly 20 amino acids – just the number life uses. So Crick’s hypothesized system was elegant, intuitive, and maximally efficient – in a word, beautiful.
And the hypothesis was wrong. If I were to design a genetic code, I’d use Crick’s system; and if biology worked like mathematics or physics, the beauty of Crick’s hypothesis would strongly suggest that it was right. But evolution didn’t use that code; it used a kludgy one, in which the 64 3-base codes are assigned to amino acids somewhat haphazardly: tryptophan has only a single code but leucine has 6, for example. And in the real code, no matter where you start, the string of bases can be read as a sequence of amino acids – so the code needs to specify a starting “frame” (ATG means either methionine or start here, which is terribly inelegant; Crick’s system didn’t need this kludge). Because evolution used this code, deletion or addition of a single base (a “frameshift” mutation) doesn’t just change one amino acid in the protein; it scrambles the entire thing. This is downright ugly***.
And yet biology is beautiful, if you know where to look. Paradoxically, “where to look” is at the very source of all the ugly kludges: evolution by natural selection. Natural selection happens whenever any set of organisms has two properties. First, heritable variation: individuals differ in ways that are passed on from parent to offspring. Second: differential survival or reproduction: that individual variation helps determine which individuals are more successful in leaving offspring behind them. With these two properties, organisms will evolve, as variants more successful at survival and reproduction come to outnumber and replace less successful variants. Natural selection has to work, given those two simple properties: it’s mathematically inevitable.
Natural selection is simple, complete, and powerful; and it has generated the spectacular diversity of life on Earth. The beautiful, the mundane, the weird, and yes, the ugly: hummingbirds, the grass in your lawn, bacteria that live in hot springs at 95 ºC, panda’s thumbs and kludgy genetic codes. All spring from the operation of a mechanism that operates inevitably and can be famously summed up in a few words (survival of the fittest). Natural selection is our Euler identity. Darwin sensed this, and ended his most famous book this way: “There is grandeur in this view of life, with its several powers, having been originally breathed by the Creator into a few forms or into one; and that, whilst this planet has gone cycling on according to the fixed law of gravity, from so simple a beginning endless forms most beautiful and most wonderful have been, and are being, evolved.”
Biology is beautiful.
© Stephen Heard (email@example.com) September 28, 2015
**^That is, the string ACTGGCTAA might be parsed …ACT GGC TAA…, …A CTG GCT AA…, or …AC TGG CTA A…; but under Crick et al’s hypothesis, only one of these readings would be interpretable, and there would be no frameshift mutations.
***^To be fair, it could be worse; the kludgy code does use redundancy to buffer the effect of base changes on the amino-acid sequence being coded for. But it could be a lot better, too.