Selecting the optimal seeds with the lowest frequency is a complex computational problem and often takes a long time. A frequency - based merged seed selection algorithm (FMSS) for reads is proposed. This algorithm can efficiently select a set of seeds close to the optimal and can be used to improve the performance of existing mapping tools. Experiments compared the average seed selection method and the current optimal seed selection strategy (OSS, optimal seed solver). The results show that the FMSS algorithm can give an optimal set of seeds close to OSS with a small time cost, indicating that the FMSS algorithm can be integrated into existing mapping tools to handle larger - scale read mapping problems.
选择具有最低频率的最优种子是一个复杂的计算问题,往往需要很长时间.提出了一种read的基于频率的合并种子选择算法(FMSS),该算法能够高效地选择接近最优的种子集合,可用于改善现有映射工具的性能.实验对比了平均种子选择方法和当前最优的种子选择策略(OSS,optimal seed solver),结果显示 FMSS算法能够用很少的时间代价给出接近OSS的最优种子集合,这表明FMSS算法可集成到现有映射工具中用于处理更大规模的read mapping问题.