Image
What is life? Information
Understanding the origin of life is perhaps the most important outstanding problem in science. Just where did life begin, what conditions does it need and how likely is it to emerge from pre-biotic soups elsewhere in the universe are questions that occupy a great many researchers.

Although fascinating questions, they are particularly hard to formulate in ways that are amenable to scientific investigation. Most research focuses on life-bearing molecules and self-supporting, or autocatalytic, chemistries. But even then, it is hard to agree even on a definition of life. So it is no surprise that progress is slow.

Today, Christoph Adami at Michigan State University in East Lansing sidesteps many of these problems by analysing the origin of life from an entirely different point of view. The basis for Adami's new approach is the idea that life is fundamentally a phenomenon of information. This allows Adami to ignore all the messy details of chemistry and instead consider life's most basic properties as ones determined by the nature of information and the laws of physics that govern it.


Comment: So far so good. If there's one thing that differentiates living organisms from basic chemistry, it is the information that codes for the materials that construct their bodies, not to mention 'epi-genetic' sources of information like body plans and cell structures.


First, some background about the nature of information and how it is related to life. Physicists have long known that information can be thought of as a measure of order. In ordinary circumstances, the amount of information always decreases over time. Or put another way, the amount of disorder (or entropy) increases until it reaches some maximum value. At this point of maximum entropy, the system is in thermodynamic equilibrium with its surroundings.

The key idea in Adami's formulation is that living systems do not exist in a state of thermodynamic equilibrium but somehow maintain themselves in a state that differs from maximum entropy by a deficit that is equivalent to the information they contain. A characteristic feature of living systems is that they can maintain this difference indefinitely.


Comment: In other words, they need a constant input of information. But what is the source of this information? What is the source of new information, i.e. new species, new traits?


Adami's approach is to put all this on a mathematical footing. He starts by assuming that a proportion of all molecules of a certain length can act as self replicators. It does not matter whether these are individual polymers like DNA or sets of autocatalytic molecules that together reproduce themselves. The key thing is that the information they contain is replicated.


Comment: Right off the bat we've got a problem. That's a big assumption, and it begs the question of how likely it is for any self-replicating system to self-evolve.


He then goes on to develop a mathematical model of the difference between the entropy of the system and a system in thermodynamic equilibrium. This difference is the information that self replicators store about what it takes to self replicate in their environment.

He then explores how the mathematical properties of this system change given that the molecules involved can occur with different frequencies and have varying characteristics themselves.

One important property turns out to be the size of the difference in entropy between a non-living system and a living system. When this difference is large, the likelihood of finding a molecule or set of molecules that can act as self replicators is vanishingly small.

But if this difference can be made small, the likelihood of finding self replicating models increases. And not by just a small amount. Adami shows that it can increase by dozens of orders of magnitude, by 64 orders of magnitude in one case. In other words, the probability of the spontaneous emergence of life varies dramatically and can be particularly favourable in certain conditions.

Indeed, he tests this idea against the data from an artificial life software system called Avida, which allows scientists to study evolutionary biology in silico. It turns out that his mathematical model accounts for the rate at which researchers have found self replicators in Avida, albeit within crude limits.

That is a fascinating approach that has significant potential for future work. The big advantage of an information-theoretic approach is that chemistry is taken out of the question. And although it assumes the existence of self-replicating polymers, these need not be chemical at all. The result is a study of the properties of that is satisfyingly mathematical.


Comment: In other words, like climate change models, they're entirely a result of the parameters created and fed into the model by the researchers themselves. They have little to do with what actually happens in nature.


It clearly has more potential. "The information-theoretic musings I have presented here should convince even the skeptics that, within an environment that produces monomers at relative ratios not too far from those found in a self-replicator, the probabilities can move very much in favour of spontaneous emergence of life," concludes Adami.

It will be interesting to see who picks up this ball and runs with it.

Ref: arxiv.org/abs/1409.0590 : Information-Theoretic Considerations Concerning the Origin of Life