Sunday, January 15, 2012

Review: First Life, by David Deamer

Just finished David Deamer's First Life.

The main point of the book is to expound on Deamer's theory that lipid based vesicles were important to the Origin Of Life (OOL). Lipids are the oils and fatty acids we use today as cell walls in our bodies. They are 'polar' molecules, with a different charge at each end of the chain of atoms. This makes one end attracted to water and the other end repelled by water (hydrophobic). When dispersed in water, the hydrophobic ends wind up crowded together, while the hydrophillic ends stick out, so the lipid molecules naturally form surfaces that are two molecules thick. Under the natural physical forces found in water, the most stable shape for these layers is a sphere -a vesicle or bubble.

A key property of these vesicles is that small molecules can work through the layer separating the inside from the environment, but a large molecule inside can't get out. Therefore, a chemical reaction that built up large molecules from small ones inside the vesicle would build up a concentration of large molecules far faster than the same reaction happening in the open, where the reaction products would get mixed and diluted quickly. It just so happens that life is full of these kind of reactions.

Having read a lot of pop sci literature (Zimmer, Ridley, etc.) it shows that Deamer is a working scientist, not a professional writer. At times the book felt padded by reviews of everything from the Big Bang onward, and an explanation of what name comes first in an article reference. And there was a fair bit of chemistry porn, in which the author gives a bit too much detail on lab procedures.

Deamer's take home message is that OOL requires some minimum complexity, and most scientists are not willing to attempt the messy experiments necessary. In a football analogy, grant funded science is a "three yards and down" ground game, and OOL needs some Hail Mary passing. 

His last chapter describes his ideal update of the famous Miller-Urey experiment, which showed that a combination of small molecules and energy could lead to biologically important molecules such as amino acids. It's big, throws a lot into the mix, and would cost a couple million dollars to run. At the same time, he acknowledges that it would have to run over and over, with multiple changes in atmosphere, temperature, pressure, etc. which would increase the cost. But at the same same time, he mentions that robotic experimentation runs hundreds of experiments at the same time. However, the two ideas never connect - that you have to reduce OOL experiments to something that can be done on a microfluidics chip in large batches.