8One of many important features of the tumour microenvironment is that it is a place of active Darwinian selection where different tumour clones become adapted to the variety of ecological niches that make up the microenvironment. These evolutionary processes turn the microenvironment into a powerful source of tumour heterogeneity and contribute to the development of drug resistance in cancer. Here, we describe a computational tool to study the ecology of the microenvironment and report results about the ecology of the tumour microenvironment and its evolutionary dynamics. 9
Introduction 10The tumour microenvironment is characterized by large chemical gradients and contains a mixture of normal and tumour cells.
11The presence of high gradients favours the formation of different ecological niches which can become an important source of 12 tumour heterogeneity 1-6 .
13The actual size of the niches is quite small, as it is mostly determined by the structure of the scaffold of capillary vessels 14 that envelope and feed the tumour mass. This structure is highly irregular, with a chaotic blood vessel network, a missing 15 lymphatic network, elevated acidity, poor oxygenation, and high interstitial fluid pressure. The distance between blood vessels 16 may be as large as a few hundred µm, and the regions in between them can become highly hypoxic 4 .
17The fine-graininess and the large spatial variability are essential features in the formation of the ecological niches and they 18 must belong to any mathematical model that aims to explain tumour heterogeneity as the result of Darwinian selection driven 19 by the local environment. The importance of the microenvironment is widely recognized, and several researchers have tackled 20 the problem of its description and understanding from the computational point of view, e.g. 7-13 , and the reviews 14, 15 .
21The studies, however, mainly focus on the biochemical, cellular and biophysical properties of the tumour microenvironment 22 and not on its active role in providing varied evolutionary paths to genetic variants of the tumour cells. As it has already been 23 pointed out 16, 17 , the adaptive evolution of tumour clones (central concept of Darwinian dynamics) is driven by the formation of 24 new environmental niches. Many practical difficulties limit the experimental study of the adaptation process, while computer 25 simulations can shed light -albeit in a limited way -on the dynamics of many steps like the convergent evolution of different 26 genotypes to the same phenotype, and the selective loss of specific cell functions.
27Simulations of avascular solid tumours show that the microenvironment of these small cell aggregates is formed by rather 28 homogeneous niches with smooth gradients of oxygen, of other nutrients, waste molecules and cell viability 18 . After the 29 angiogenic transition, however, the microenvironment differentiates in unpredictable ways. To identify underlying processes and 30 provide reasoning, we have developed a computational model that is a tool t...