Abstract
The behavior of natural, biological ecosystems is for a large part determined by environmental conditions. It should therefore be possible to experimentally manipulate such conditions to drive ecosystems in desirable directions. When a set of environmental conditions can be manipulated to be either present or absent, such an exercise becomes a typical combinatorial optimization problem, and one for which a genetic algorithm should be very suitable. In this work, four exhaustive experimental datasets were assembled, containing growth levels of different natural microbial ecosystem as influenced by all possible combinations of a set of five chemical supplements. The ability of a genetic algorithm to search these datasets for combinations of supplements driving the ecosystems to high levels of growth was compared to that of a random search, a local search, and a hill-climbing algorithm. The results indicate that a genetic algorithm is very suitable for driving microbial ecosystems in desirable directions, which opens opportunities for both applied and fundamental ecology.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have