Simulation of organisms using a developmental model part 1: Basic description

Abstract

We discuss a program for simulating linear arrays of cells, each one changing under the same set of rules. The action of a cell is influenced by both its neighbours. Thus, the program can be used for the simulation of linear iterative networks of identical modules. The authors' main application, however, is in the field of biology, where the program can be used to test hypotheses about the developmental rules for organisms. For this reason, the division of one cell into several cells is allowed. If developmental rules based on the hypotheses are fed in as data to the program, it will give as output the case-history of any desired number of developmental sequences, as well as statistical data concerning the development. These can then be checked against experimental results and thus the validity of the hypotheses can be established. For well-validated hypotheses the program can also be used as a predictor. The advantage of using the program described here over writing a simulator for each specific biological situation is the great reduction in programming time. A user of our program need only write a simple routine to describe the local interaction between cells and the kind of statistical data he requires, and the program will automatically work out the global behaviour of the organism and collect and print out the desired data in an acceptable form. Even a very inexperienced programmer can use our program easily and effectively for biological simulation. The program is demonstrated by applications concerning the developmental rules for red algae and for molluscan pigmentation patterns. In Part 2, we shall investigate some hypotheses concerning heterocyst formation in blue-green algae.