PerleDeDiamant

PerleDeDiamant

Why Life Is Physics, Not Chemistry

Why Life Is Physics, Not Chemistry

The idea that life boils down to chemistry is being usurped by a much more ambitious idea, says two of the world's leading biophysicists

kfc 11/22/2010

 

In the history of science, there are many examples of simple changes in perspective that lead to profound insights into the nature of the cosmos. The invention of the telescope is perhaps one example. Another is the realisation that chemical energy, thermodynamic energy, kinetic energy and the like are all manifestations of the same stuff. You can surely supply your own favourite instances here.

One of the more important examples in 20th centruy science is that biology is the result of evolution, not the other way round. By that way if thinking, evolution is a process, an algorithm even; albeit one with unimaginable power. Exploit evolution and there is little you cannot achieve.

In recent years, computer scientists have begun to exploit evolution's amazing power. One thing they have experienced time and time again is evolution's blind progress. Put a genetic algorithm to work and it will explore the evolutionary landscape, looking for local minima. When it finds one, there is no knowing whether it is the best possible solution of whether it sits within touching distance of an evolutionary abyss that represents a solution of an entirely different order of magnitude.

That hints at the possibility that life as it has evolved on Earth is but a local minima in a vast landscape of evolutionary possibilities. If that's the case, biologists are studying a pitifully small fraction of something bigger. Much bigger.

Today, we get an important insight into this state of affairs thanks to a fascinating paper by Nigel Goldenfeld and Carl Woese at the University of Illinois. Goldenfeld is a physicist by training while Woese, also a physicist, is one of the great revolutionary figures in biology. In the 1970s, he defined a new kingdom of life, the Archae, and developed a theory of the origin of life called the RNA world hypothesis, which has gained much fame or notoriety depending on your viewpoint.

Together they suggest that biologists need to think about their field in a radical new way: as a branch of condensed matter physics. Their basic conjecture is that life is an emergent phenomena that occurs in systems that are far out of equilibrium. If you accept this premise, then two questions immediately arise: what laws describe such systems and how are we to get at them.

Goldenfeld and Woese say that biologists' closed way of thinking on this topic is embodied by the phrase: all life is chemistry. Nothing could be further from the truth, they say.

They have an interesting analogy to help press their case: the example of superconductivity. It would be easy to look at superconductivity and imagine that it can be fully explained by the properties of electrons as they transfer in and out of the outer atomic orbitals. You might go further and say that superconductivity is all atoms and chemistry.

And yet the real explanation is much more interesting and profound. It turns out that many of the problems of superconductivity are explained by a theory which describes the relationship between electromagnetic fields and long range order. When the symmetry in this relationship breaks down, the result is superconductivity.

And it doesn't just happen in materials on Earth. This kind of symmetry breaking emerges in other exotic places such as the cores of quark stars. Superconductivity is an emergent phenomenon and has little to do with the behaviour of atoms. A chemist would be flabbergasted.

According to Goldenfeld and Woese, life is like superconductivity. It is an emergent phenomenon and we need to understand the fundamental laws of physics that govern its behaviour. Consequently, only a discipline akin to physics can reveal such laws and biology as it is practised today does not fall into this category.

That's a brave and provocative idea that may not come as a complete surprise to the latest generation of biophysicists. For the others, it should be a call to arms.

We'll be watching the results with interest.

Ref: arxiv.org/abs/1011.4125: Life Is Physics: Evolution As A Collective Phenomenon Far From Equilibrium

 

Torbjorn_Larsson_OM No contender to evolution

Well, call me underwhelmed.

Yes, Woese is a known name, and while I haven't studied the history of the RNA world (several players of note) it is indeed a famous and rather well tested theory; a chicken-egg problem of the DNA and protein type in translation points to a rather restricted bottleneck of pathways and the RNA machinery in its core was found to be it.

(There is a "physical" principle in biology, by the way.)

But he is also known for the vertical gene transfer meltdown consortium idea that was hypothesized to replace a LUCA, or at least an LCA, something that seems to be rejected by modern whole genome phylogenies. ("The mega-matrix tree of life: using genome-scale horizontal gene transfer and sequence evolution data as information about the vertical history of life", Lienau et al, Cladistics 26 (2010) 1–11.) Something that Goldenfield and Woese still worked on 2007, according to the arxiv.

Anyway, this review does nothing to replace evolution as the basic theory and process of life. It is as far as I can see from a quick read merely handwaving for "the conceptual failings of the modern evolutionary synthesis". Which is rather ridiculous considering that the synthesis is the best tested theory in science (due to its concrete complexity and many mechanisms).

As for the proposals themselves, an example of a simple problem with the paper, take "the far from equilibrium" idea. This is nowhere to be seen. Life is working at the order of chemical potentials driven by light. There is then no difference between the surface of Earth and cells with respect to their idealized steady state and its placement relative to a thermodynamical equilibrium. And no one goes around claiming that there should be revolutionary "physical principles" involved in explaining weather or climate.

As for evolution itself, there is a paper on arxiv ( http://www.arxiv.com/abs/0903.4603 ) that puts the entropy production from evolution ~ 10^-10 of the entropy production from Earth's absorption of sunlight.

In fact, proposing "physical principles" instead of biological theory in biology is a popular if misdirected pastime which have no specific reason to succeed, and is seen to fail constantly. It is a field of mostly cranks and crackpots. I'm not tempted to dig deeper into the paper.

However, a nitpick on the post: evolution is a local hillclimber, it isn't explicitly looking for local minimums.

Indeed, it is trivial from population genetics that to find a local minimum, assuming that it can reach it with modifying alleles, it will take an infinite population infinite time under "variation and selection". And the phenomena of near neutral drift assures us that it will never get there except by unlikely coincidence, and may as well drift away from a minimum in a real finite population!



22/11/2010

A découvrir aussi


Inscrivez-vous au blog

Soyez prévenu par email des prochaines mises à jour

Rejoignez les 3 autres membres