A framework for cancer-related genes mining over the Internet

Clinically, cancer is a complex family of diseases. From the view of molecular biology, cancer is a genetic disease resulting from abnormal gene expression. This alternation of gene expression could be resulting from DNA instability, such as translocation, amplification, deletion or point mutations. A large amplification or deletion of a chromosome region can be easily detected by two methods: loss of heterozygosity (LOH) and comparative genomic hybridization (CGH). The different gene expression pattern can be monitored by high throughput microarray analysis. Enormous data accumulated by practicing these technologies and the data pool is continuing enlarging with an amazing rate. To aid investigators mining useful information in these data deposits, new data storing and analysis tools must be developed. Two value-added databases are constructed to achieve this purpose. They contain information of genes in the unstable regions of cancer cells basing on the data accumulated from LOH and CGH experiments and information of cancer cell gene expression profiles according to microarray analysis, respectively. An automatic system to retrieve interesting gene information, to compare with the known databases, to analyze and predict the protein functions, and to group the genes of the same function will be integrated into the database circuit. An automatic update system will be installed and performed after the setup of the two databases. The system keeps also the probability to modify and to accept new data obtained from any new techniques. Our goal is to help biologists to find the needles in a haystack that is, to find the real cancer-related genes (oncogenes or tumor suppressor genes) for further research purpose.