A version of this applet using the native VM is available here.

The Evolution of Sex

The model presents a field of organisms which reproduce either sexually (green organisms) or asexually (blue organisms).

The Red Queen - pressure from parasites

Organisms are modelled as a collection of genes for disease resistance - parasites have to match the 'locks' of their hosts to a specified degree in order to be able to reproduce.

Gene repair - coping with genetic error

Organisms with more than this critical number of errors turn red and slowly die.

In this model, whether sex evolves depends primarily on the mutation rate. With a low mutation rate, asexual creatures thrive. Increasing the mutation rate causes sexual creatures to do better.

Note that increasing the mutation rate to its maximum level usually has the effect of extinguishing all life, though.

Note that the gene repair model is capable of overcoming the "two-fold cost of sex" (which may be imposed artificially by unticking the "internal fertilisation" checkbox).

One interesting aspect of the model is that, in an environment where sexual organisms usually have a slight advantage, asexual organisms may often be produced in significant numbers buy simply manually clearing a large area of the grid and allowing organisms to expand into it.

In nature a number of organisms reproduce asexually when environmental selection pressures are reduced. Our model suggests that this may sometimes be a natural outcome of natural selective processes, rather than a response genetically programmed into the organisms involved.

General points

The applet conclusively illustrates a number of cases where the advantages of sexual recombination are sufficient to drive the evolution of sexuality in such a population.

Sexual organisms are modelled as internally fertilising hermaphrodites. This means that they need invest little in producing male adaptations as elaborate adaptations make no difference to the success rate of seed.

As far as we are aware, the model here is the first to demonstrate the ability of sexual organisms to successfully invade an asexual population. No doubt this claim will be rapidly be withdrawn when (for example) William Hamilton's second volume of collected papers finds its way here.

Another interesting feature of the model is that there is a switch controlling whether the host organisms have haploid or diploid genomes.

The total size of the genome remains unchanged by this switch. Diploid populations display significantly better parasite resistance than haploid ones, and it is in diploid populations that the evolution of sex is most readliy observed.

In the biological literature there are a large number of suggested functions for diploidity. It has been satisfying to find a concrete model which indicates that there is a significant advantage to diploidity in a simple model of parasitism.

The model also exhibits a number of other features: large scale spatial patterns occur and phenomena such as population oscillations may be observed.

Future plans

A more detailed study of how haploid organisms compete against diploid organisms in response to selection pressure from parasites appears to be indicated.

We have made no model of sibling-rivalry pressures or "Vicar of Bray" mechanisms which may contribute to the maintenance of sex. Though we think these may be of significant importance under some circumstances, their role in the origin of sex appears to be low, and it appears unlikely that such effects will be of significant relevance to our main project. Consequently, we have no plans to model these effects at this stage.

Our model deals with the origin of sexual recombination, not the origin of males and females. There are a number of theories which suggest important roles for sexual dimorphism - and at least one of these is amenable to computer simulation. These theories are of some interest on theoretical grounds - but again, at this stage are largely periphereal to our main project.

• Model type - illustrate parasite pressure or gene repair theories;
• The rate of cell growth - controls how fast healthy organisms grow;
• The rate of disease spread - controls how rapidly parasites reproduce;
• The rate of cell death - controls how fast infected organisms die - slow death of infected hosts produces larger scale patterns, but it also increases the chance that the parasites will wipe out small populations completely;
• The rate of infection - controls how closely host genes and parasite genes need to match before parasites can reproduce;
• Allow reinfection - if ticked, infected cells may be reinfected;
• Diploid hosts - if ticked, host cells use diploid recombination. Use of a diploid crossover operator appears to magnify the advantage of sex.
• Internal fertilisation - if unticked, sexual organisms are penalised by reducing their fertility by a factor of two. This currently prevents sexual organisms from competing effectively.
• Display - controls which aspect of the automata is presented:
  • Species - infected organisms are coloured by the 'strain' of the infection;
  • Health - infected organisms are coloured according to their health;
• Mutation rate - controls how rapidly the genes for parasite resistance in the host species change with time;

This applet can also be run as an application. Download this jar file (using shift-click) and double-click on it.