333

More details can be found in the paper (including details on the semiquantitative simu­

lation of the different system states): Brietz A et al. (2016) Analyzing ERK 1/2 signalling

and targets. Mol Biosyst.

20.10 Understand Evolution Better Applying the Computer

Question 10.1

Evolution is the change in characteristics of living organisms over time. Important mecha­

nisms are, for example, mutations, selection, gene drift and separation.

Questions 10.2 and 10.3

One color always prevails in the end. We have a Darwinian evolutionary approach here.

The probability of being hit is directly proportional to the number of individuals. Random

fluctuations, however, lead to the random extinction of individual colors until eventually

only one color remains. This “game” vividly reproduces neutral evolution (all colors have

exactly the same chance of winning at the start, and as colors become fewer, their die-off

rate becomes proportionally lower). So just pure fluctuation, and yet one color eventually

prevails. This simulates genetic drift very nicely.

Of course, it is also very easy to simulate selection for the “fitter” by modifying the

rules of the game, e.g. that one color (red) simply gets two offspring for each hit and you

always randomly roll two individuals for this case. Then red always wins. How fast this

happens depends on randomness. So the result here is predictable, but the sequence of the

individual steps is not.

True evolution is always a mixture of both, lots of drift involved, as perfectly illustrated

in Stephen Jay Gould’s “A wonderful life”.

Question 10.4

Now the probability that a tandem of two colors asserts itself is proportional to the product

of both colors. Thus, the more individuals there are for a tandem, the quadratically better

the rates. This is why a “once and for all” selection occurs. Quite quickly a tandem of two

colors asserts itself, and no other tandem can grow so high, because no population can

compete with the super-exponential reproduction rate.

This simulation model nicely illustrates how over-exponential growth prior to the first,

delimited cells led to selection from a population of mutually catalyzing molecules. In

particular, it explains very well why only one genetic code (with minimal dialects)

remained.

Additional task for those interested: write R code to recreate the three games (not dif­

ficult, but takes some time).

20.10  Understand Evolution Better Applying the Computer