基于统一投影和邻居桶聚集提炼策略的基序查找算法

基序查找是生物信息学中的一个重要问题,由于生物序列中大多数信号的复杂性,一直没有很好的模型或可靠的算法来求解这一问题.本文提出了一种基于统一投影和邻居桶聚集提炼策略的基序查找算法UPNT(Uniform Projection with Neighbourhood Thresholding).在UPNT算法中,利用统一投影策略有效减少了投影数目,并使用邻居桶聚集提炼的策略大大减少了提炼桶的数目.本文进一步使用背景分布均衡与非均衡的合成(l,d)序列两套数据集对算法性能进行测试和分析,实验结果表明:UPNT在成功率和运行时间上的综合性能优于Random Projection、Aggregation和Uniform Projection等投影算法,具有更强的适用性.

Motif Finding Algorithm Based on Strategies of Uniform Projection and Neighbourhood Thresholding

Motif finding is a significant problem in Bioinformatics,while according to the complexity of most signals in biologic sequences;there aren't any extremely good models or dependable algorithms to solve this problem.This paper introduces the UPNT(Uniform Projection with Neighbourhood Thresholding) algorithm,which based on,two efficient strategies,Uniform Projection and Neighbourhood-based Thresholding.In UPNT algorithm,the policy of uniform projection leads to less projections,while the strategy of refining the buckets after aggregation results in great abatement of the number of buckets to be refined.This paper further demonstrates the test results and analyses of UPNT algorithm and other algorithms experimented on two synthetic(l,d) datasets with unbiased as well as biased background,while the results demonstrate: UPNT algorithm shows superior to Random Projection,Aggregation and Uniform Projection in the synthetic performance of success rate and average running time,therefore carries greater applicability.