%A WU Huimin, WU Jingli %T An improved cycle basis algorithm for haplotyping a diploid individual %0 Journal Article %D 2016 %J Journal of Shandong University(Engineering Science) %R 10.6040/j.issn.1672-3961.1.2016.229 %P 9-14 %V 46 %N 4 %U {http://gxbwk.njournal.sdu.edu.cn/CN/abstract/article_1461.shtml} %8 2016-08-20 %X HapCompass is an effective heuristic algorithm for solving the minimum weighted edge removal(MWER)model. The HapCompass algorithm eliminated conflicting cycle basis by deleting the edge with the minimum absolute value of weight. When there were several edges whose absolute values of weights were equal to the minimum one, HapCompass chose the deleted one randomly, which led to produce uncertain solutions and decrease the reconstruction effect. Aiming at this problem, algorithm IHapCompass was proposed. It improved the rules of deleting edges to limit the random value effectively. IHapCompass took advantage of the relativity between the difference of fragment number of(00)/(11)and that of(01)/(10)and the total fragment number to ascertain the deleted edge. It made use of the probability of 0/1 in haplotypes to determine the SNP value for an isolated vertex, which ascertained the value of an isolated vertex. Experiments were implemented by using the real haplotypes. The results showed that under different parameter settings, the IHapCompass algorithm can obtain higher reconstruction rate than the HapCompass, the DGS and the Fast Hare algorithm, and has high efficiency. The IHapCompass algorithm could provide a reference for the study on reconstructing single diploid individual haplotypes.