Fast Decision Tree To Index Large DNA-Protein Sequence Datasets

The significant growth in biological data has motivated important research by computer scientists into creating fast and accurate algorithms for accessing the database at the fastest possible rates. Therefore, there still remains the need for an efficient indexing approach to speed up the searching time for large biological data, to help the biologists to retrieve the exiting information quickly, from any biological database. Furthermore, the trends in parallel programming models has also encouraged the computer researchers to improve the indexing approaches to cope with this exponential increase. This book adopted a Decision Tree method as an Indexing Model (DTIM), to allow for large databases to be processed. This method of indexing could effectively and rapidly retrieve all similar DNA-Proteins data from a large database for a given query. The decision tree indexing model (PDTIM) was then parallelized, using a hybrid of distributed and shared memory to accelerate the index building time. Finally, to improve the accuracy rate for decision tree algorithm, the decision tree was hybridized with a harmony search algorithm (HDT-HS).