Tumor Growth Model

There are a number of biological processes that are amenable to an agent-based approach. One of these processes is the formation of tumors, commonly formed by self-organizing cancer cells.

We present a Simudyne implementation of the following publication: An Agent-Based Model of Avascular Tumor Growth

The code for this model is available as a zip download, containing a complete model as a Maven project. This project should be able to be run from any Maven or Java IDE environment.

Abstract

We developed a simplified agent-based model for an avascularised tumour. The model takes into account a healthy tissue in which blood vessels introduce nutrients that diffuse. In this tissue, cells move/proliferate/die according to their energy and the available space for their offspring. They can mutate to cancerous with a certain probability and can also change mutated which means that they are affected by their neighbours states. First, we describe briefly the background of the problem and previous models made for the particular avascular state. Then, we describe in detail the rules and actors our model takes into account and comment on the particular choices. Next, we present our results from the parameter exploration. We were interested in finding the critical values that define the transition between a majority of cancerous cells and normal cells. We studied the outcome of varying the effect of neighbours, probability of division and mutation probability. We discovered a phase transition in the effect of neighbours parameter space and the effects of the other parameters in this space. Finally, we discuss the relevance of the model as well improvements one could make.