RS-GA-KNN算法识别灵长类动物DNA序列剪接位点

以灵长类动物DNA序列的剪接位点识别资料为研究对象，将选定样本序列中各碱基编码作为原始变量数据，用粗糙集方法和遗传算法对原始变量数据进行变量筛选，即以粗糙集方法选取的变量为基础，用遗传算法进行变量的二次搜索，从样本序列各碱基中挑选出保守性强的碱基对应的变量构成变量集，采用最近邻聚类识别灵长类动物DNA序列剪接位点类型，总识别准确率达90．66％，明显高于直接使用原始变量数据或将粗糙集理论方法和遗传算法单独用于变量选取的识别结果．

Rough set-genetic algorithms-k-nearest neighbor and its application in DNA sequence splice sites recognition

Rough sets theory is used to deal with the data of primate DNA sequence splice sites in this paper. The variables of data are selected. The genetic algorithms is modified by rough sets and coded as rough set genetic algorithms based on the rough sets theory result. The solution is searched for the second times by genetic algorithms. The rough set genetic algorithms answer space is efficiently restricted,and its efficience of getting answer is strengthed. Rough set-genetic algorithm is likely to get the useful information which is deleted by rough sets theory,to advance the algorithms the ability of searching the answer. The correctness percent reaches 90. 66%,which is obviously better than other methods got the results.