Integrative genomic analysis of sporadic colorectal cancer.

摘要

Colorectal cancer is one of the leading causes of cancer deaths in the westernized world. The great majority of these cases are sporadic occurrences with a familial genetic component reported. Multiple susceptibility alleles, dietary, and environmental risk factors combine to determine individual predisposition. Attempts to map colon cancer susceptibility alleles in the human population have proven difficult due to failure to achieve statistical significance or have resulted in the identification of several weak alleles with limited clinical significance. However, a number of low penetrance susceptibility alleles have been identified in mouse studies of segregating populations. We propose the use of a more diverse backcross population to identify additional susceptibility to colon cancer (Scc) modifiers as current loci have been realized with limited genetic diversity on common inbred strains.;With the inclusion of the Mus spretus strain, the azoxymethane (AOM) model of sporadic human colorectal cancer mimics the heterogeneity, pathology, and molecular changes evident in human populations. In contrast to previous studies, we seek to expand on identify modifying loci and wish to extend our knowledge to reconstruct cancer specific networks and the interactions occurring therewith. Using classical genetic mapping techniques we can begin to visage the location of genes controlling the most variance, or strong effect modifiers. Two-dimensional epistatic scans enable us to identify additional small effect modifying loci and begin the loose network recreation. These results are combined with a systems biology approach, linking transcript expression with genomic variation allowing us to identify the major components of a susceptibility network and guide candidate gene selection. We conclude that these efforts will assist in susceptible patient detection and pharmacogenetic targeting of the most critical risk alleles.