Cancer is in part an evolutionary problem. Tumors are the result of genetic mutations conferring selective advantage to individual cells. Cells with selective advantage eventually dominate the tissue. This is the basis for both the virulence of cancer and our difficulties in treating it. In order to understand this aspect of cancer, our work has primarily focused on computational modeling of the evolutionary system.

Grasping the dynamics of cancer has proven to be difficult for cancer researchers. Representing heterogeneous populations of agents (i.e. cells) interacting in the spatially structured environment becomes problematic with tractable mathematics. Computational models can help to extend analytical theory to the dynamics of such systems. Our models currently focus on examining the role of neutral and selective mutations in tumor development (Maley & Forrest, 2000) and on the hallmark characteristics of cancerous cells (work by Robert Abbott).

The progression of these mutations in cancer tissue can be determined by reconstructing the phylogenetic tree. In conjunction with Carlo Maley, we developed techniques to reconstruct phylogenies of mutant clones in the neoplastic tissue of Barrett's esophagus patients (Maley et al., 2000).