In our previous discussions on how nature handles information we said that whatever system we use to store our biological information is going to have to be:
- Small
- Complex
- Stable
- Accessible
- Abundant
In this post let’s dig into that first criterion – small.
I will be aided today by three of the 20th centuries greatest scientists (and scientific communicators, no less): Richard Feynman, J. B. S. Haldane, and Erwin Schrodinger. This is not the start of joke, it is a serious discussion. Promise.
I will draw heavily from Feynman’s talk: Plenty of Room at the Bottom, Haldane’s essay: On Being the Right Size, and Schrodinger’s book: What is Life?.
Why are Molecules so Small?
In What is Life? Schrodinger poses an interesting question – why are molecules so small? It got me thinking “yeah, why are they so small? Bloody things. Well, come on, surely they have to be.” “They are made up of atoms” I mumble, “the atoms are separated via Lienard-Wiechart potentials” I say, as I wave my hands, losing confidence. It is not a satisfying answer. It does not explain anything.
Luckily, Schrodinger bails me out. He goes on to ask the much more interesting question – why are molecules so small compared to us? We are interested in molecules because of the role they play in supporting life, so why do they have to be so much smaller than the lifeforms that they enable? Now that is a really good question.
The answer he puts forth (at least in the first instance) is “because statistical mechanics” (not actually a direct quote). Statistical mechanics basically says such and such is happening because you have a huge number of individual particles. If you look at them too closely then you will see they are going nuts, but if you look at the “big” picture – the ensemble of those particles – then they do seem to be obeying physical laws. For example, is it valid to say that a single molecule is diffusing? Certainly it wouldn’t be capable of obeying Fick’s Law, for example.
So in a sense, life needs some sort of laws that it can abide by and depend on. To do that, there must be loads of molecules present such that the classical view of statistical mechanics can apply, and so lifeforms have to be much larger than molecules. Simple. Some lifeforms can get so much bigger than molecules that they even start asking misguided questions like “why are molecules so small?”, like it is some sort of failing on the part of the molecules. It isn’t.
It’s About Being the Right Size
Enter Haldane. On Being the Right Size is probably one of the most entertaining papers that you could ever hope to read. It primarily addresses the differences in size of different animal species:
“For every type of animal there is a most convenient size, and a change in size inevitably carries with it a change of form”.
Haldane, On Being the Right Size
I think this is a nice lens to view different situations through, however. For example, as organisations grow in size they invariably need to change structure, usually due to the increasing strain being placed on the existing communication networks. Or consider the situation where you are walking to meet friends and you realise you are running late. What do you do? You cannot simply keep increasing your walking speed; there comes a point where you are forced to change gait, either to a jog or to run. Okay, you could try that Olympic speed walking thing but only if you really wanted to look weird and you were keen to do long-term damage to your hips.
Haldane gives the example of the Gazelle. What should it do if it wants to become larger without breaking its graceful little legs due to an increase in bulk? It has two choices. It can follow the rhino model or it can follow the giraffe model (both members of the same phylogenic order as the gazelle). In the rhino model you make your legs shorter and thicker, so that your increased weight has the same area of bone to support it. In the giraffe model you compress your body and stretch your legs outwards as much as possible, increasing your stability.
It is worth noting that both rhinos and giraffes run faster than humans, and they don’t go in for all that hip-shattering fast walking nonsense either, thank you very much.
In short, the physics you must concern yourself with is entirely dependant on the scale that you are operating at. Insects can easily survive 1000-feet falls, whereas I could not; gravity is of absolutely no consequence to insects. On the other hand they are probably deathly afraid of surface tension. I can easily exit a swimming pool or have a drink of water (humble brag), but for an insect these are life-threatening scenarios (insects have proboscis precisely so they can drink water without standing too close to it!). Haldane makes the wonderful comparison that drinking water is as dangerous to an insect as it would be for a man to reach out over the edge of a cliff for food.
In biotechnology we face this problem all the time. You have studied some system “in the bulk” and you think you can make good predictions, but then you deploy to a microfluidic environment. You likely find yourself in a strange new diffusion-dominated world, your Reynolds number having changed by a few orders of magnitude. Scale matters. And if you want to do engineering at microscopic scales then you really need to pay attention.
Engineering at Small Scales – Plenty of Room at the Bottom
In his talk, Plenty of Room at the Bottom, Feynman discusses engineering at the nanoscopic scale. He talks about writing the Encyclopaedia Brittanica on the head of a pin and miniaturising computers (he basically discusses integrated circuits and microchips – and this is in the era where germanium is still being used to make transistors!).
Hang on, this sounds like we would have a good way of writing information at the scales that could, in principle, be used by life. Feynman’s idea here is basically to use something like morse code and pattern a material through the use of two different metals: one for the dots and dashes, and another one for the spaces. You might have some issues with metal diffusion at the interfaces, so to preserve the fidelity of your encoded information Feynman recommends each dot is made up of roughly 100 atoms. That is an information density of 100 atoms / bit. Pretty wild. With that you could build up a patterned cube of material that contains all the information we have ever written down in books (late 1950’s). That cube would be the size of a spec of dust. Just think about that.
Admittedly there are a few draw backs to that. If you wanted to read that information back out it is not obvious how you would do it. What sort of machine would you need in order to let you read the information buried in the interior of the cube? So in the end even this extreme engineering solution wouldn’t be as good as DNA. And that is before you consider the informational density of DNA itself.
Because you have to use multiple atoms per bit in the patterned-metal approach you cannot really hit the “ideal” situation of one atom per bit. DNA actually beats Feynman’s patterned cube. To store one bit of information using DNA you need only 50 atoms – it is twice as good!
Back to Erwin
Here is the paradox as I see it. For biological machines to develop they somehow need to “obey” the laws of physics. But if we look at some of the arguments that come out of statistical mechanics, you basically need an astronomical number of molecules before these classical laws start to apply. But somehow nature is able to guarantee the persistence of biological information using only fifty atoms per bit! What is going on?
It is so poetic (though perhaps not surprising) that it is through Schrodinger that we get gain some insight into this. The answer, ultimately, is “because quantum mechanics” (again, not a direct quote). Classical physics alone completely fails to explain what life is or how it is possible. Once you get down to the single molecule level it is all quantum mechanics. It plays a role in hereditary; jump mutations in chromosomes are directly analogous to quantum mechanical jumps between energy levels.
But this all just raises more questions for me. We are satisfied that classical physics cannot explain life. But can quantum mechanics? Are we back where we started, largely agreeing that physicists are annoying? I’m not sure, but I am interested to keep exploring. But it does leave me reflecting on a few funny observations on the differences between biologists and physicists.
Biologists and Physicists
In his talk, Feynman makes an observation that ties in closely with my own experiences on how biologists and physicists see each other (part of the reason that I am tracking “biologist annoyance” throughout the DNA sonata project).
Physicists tend to be strong generalisers. They are good at making models and telling themselves that they can understand any system, in principle. They open themselves up to the usual parody “oh yeah, spherical cows in a vacuum is it?”, but all good physicists know that their models are wrong. But they might just be useful. My observation of biologists is that they are, actually, just a little bit busy handling the thousands of details that impact the reality of the situation at hand. This excerpt from Feynman is too good not to include in full (the parenthesis are Feynman’s too):
“We have friends in other fields – in biology, for instance. We physicists often look at them and say, “You know the reason you fellows are making so little progress?” (Actually I don’t know any field where they are making more rapid progress than they are in biology today.) “You should use more mathematics, like we do.” They could answer us – but they’re polite, so I’ll answer for them: “What you should do in order for us to make more rapid progress is to make the electron microscope 100 times better.”
Feynman, Plenty of Room at the Bottom
It brings to mind conversations we have a lot in the world of multi-disciplinary biotech startups. What is a product requirement and what is a specification? Can you give me this solution to this problem please versus I have this problem, could you solve it please?
That is a topic for a future post but it is funny to see that those challenges in communication have not changed much in the sixty years since Feynman spoke about them.