Estimation of degree of dissimilarity between DNA sequence Using Hellinger distance

• Main Authors: Soa-Yu Chan, 陳秀瑜
• Other Authors: Tiee-Jian Wu
• Format: Others
• Language: en_US
• Published: 2006
• Online Access: http://ndltd.ncl.edu.tw/handle/74665426221949756157

碩士 === 國立成功大學 === 統計學系碩博士班 === 94 === In molecular biology, the issue of quantifying the similarity between two biological sequences is very important. Several measures of DNA sequence dissimilarity have been developed in the past. The purpose of this thesis is twofold. Firstly, we use large scale simulation to compare the performance of Hellinger distance (HD) and symmetric Kullback-Leibler discrepancy (SK-LD). Secondly, we compare the actual CPU time and memory space required on a PC in computing HD and SK-LD. Our simulation study and real data analysis show that (1) the performance of HD is almost as good as SK-LD and (2) the computational efficiency of HD, in terms of CPU time and memory space required, is significantly better than those of SK-LD.