Mining Association Rules In DNA Database

After the Human Genome Project was completed in 2003, the era of post Human Genome is coming. Various post genome projects are being planned and putting into practice, which has sparked off vast amount of biological molecular data. To exploit such invaluable data and discover inherent biological information from the data is from all aspects important and of scientific value for human being. Researchers faced with the challenge of addressing related works are devoting all of their efforts to the studies of uncovering biology secrets from the data. Up to now, many meaningful methods and tools have been developed and implemented for manipulating genome data, however, the results of discovered patterns and prediction turn to be out of the expects. Hence, researchers still have a long way to go.Based on the frame of support-match, in this paper, the author improved the original algorithm for mining association rules, and proposed an enhanced algorithm for mining closed frequent patterns from DNA sequences. RAK alpha gene and HBsAg gene of DNA sequence data from NCBI were used in the paper. The results of analysis show that the proposed method is capable of finding all patterns of DNA sequences fast and accurately, as well as the inherent biological linkage among these patterns. Furthermore, association rules mined from the patterns can be employed for classifying different types of genes and consequently for distinguishing different types of diseases.