How did genes evolve?

A long standing problem I have had about Richard Dawkins' claim that "life began when a molecule gained the ability to replicate itself" is, how did that happen? If evolutionary biology is about remove the miraculous from explanations of life and its past, then why posit a miracle of that kind?

Some properties of the chemicals involved had to lead up to the ability to self-replicate. I have a story of my own, which I'll add at the end, but for now, Science is reporting that ribozymes have a feature which will allow them to copy RNA reliably even if there are major mutations made to the molecules.

Here's the difficulty. Without some method of error correction, such as the ones that DNA replication systems have, a genome of RNA can only get so large before mutation rates, which are a relatively constant frequency, render the molecule too error-prone to act as the genetic molecule in a living system. This has been called by Eors Szathmary the "error catastrophe", and it has been a real stumbling block for scenarios of how self-replication could get going in the first instance.

Now Szathmary and his collegues Adam Kun and Mauro Santos, working in Budapest, which is a real centre for Origins of Life research for some reason, have shown in a paper in Nature Genetics that ribosomes, which are the "machines" (it's a metaphor, OK? No ID implications) of nucleotide replication, can withstand a massive amount of sequence change before they cease to do their work. This means that even if mutations sneak into the RNA genome, the ribozymes will still replicate RNA, thus cutting off the error catastrophe. This is a major achievement (and another blow to the God of the gaps devotees). Now we have the capacity to generate genomes up to the size of a transfer RNA in a modern cell, and that is no small thing, forgive the pun.

But it leaves us wondering how genetic fidelity improved. And the answer to that question was, I believe, inadvertently given byNobel Laureate Manfred Eigen in the 1970s. He and his colleague Peter Schuster proposed a closed cycle of chemical reactions, where each products of one step catalysed the next, and called it a Hypercycle. Now suppose the following scenario for the very beginnings of organic life, before there were even cells.

There is a chamber, say an underground volcanic-water fed one, in which the source molecules from which life is made are being fed. It's hot enough to make chemical reactions occur (let's suppose there's a chemical gradient and a thermal gradient), but not so hot that the products denature. Chemical reactions might set up hypercycles.

Now it follows that the physical properties of the molecules involved in a hypercycle will vary in their stability, how well and accurately they catalyse the next step in the cycle, and how much energy they take. There will tend to be chemical selection, therefore, for reaction cycles that have at least one more stable (that is, long-lived), accurate and cheap molecule in the hypercycle.

So our proto-RNA/ribozyme complex will become optimised over time. And this allows genes to evolve. And it's non-miraculous. Straight Darwinian selection, without replicators.

This undercuts the necessity for there to even be replicators before we get a selection process going, and so it has some implications for cultural as well as biological and protobiological evolution.

Some thoughts..., and the hard work on this was done not by me but by Clem Stanyon and Ian Musgrave, in case anyone wants to credit it.