基于最小点覆盖及多参数方法的关键蛋白识别

针对已有方法对关键蛋白识别度不高的现状，认为进一步提高识别度有两条途径：一是发现与关键蛋白关系更密切的参数，二是充分挖掘现有参数的信息并进行有效地整合。由于点覆盖在网络（图）拓扑结构上的重要地位而研究将其引入关键蛋白质的识别中：针对算法的复杂性引进参数计算的相关算法将复杂度大幅度降低的同时对蛋白质网络进行最小点覆盖分析并获得一种新的拓扑参数-点覆盖参数，相关分析表明该参数与关键蛋白有着密切的联系。进一步研究发现，参数之间相关性的大小在很大程度上预示它们所蕴含的关键蛋白信息之间互补性的强弱，根据这一发现探讨利用包括点覆盖在内的各个参数的有限信息进行有效整合，仿真结果证实该方法能明显提高关键蛋白识别度。

On Essential Protein identification based on Minimal Vertex Cover and multi-parameter

Directed at the dissatisfaction status quo in the identification of Essential Protein(EP),two ways are put forward to improve the performance:one is to find out new parameters closer to EP,the other is to integrate some of known parameters.For its important role in the topology of networks,the Minimal Vertex Cover(MVC) is introduced into the study of EP in a protein network first time:the use of the algorithms relation to parameterized computing reduces the complexity of MVC problem of a protein network to a great extent so as to analyze the problem quickly and get a new topological parameter-Vertex Cover Parameter(VCP),with its closer relation to the essentiality of a protein.In-depth study,the degree of the relativity between known parameters,including VCP,is found out,which indicates the extent of the mutual complement of information of EP their containing.The integration of the information of the parameters is explored with this observation,and the identification ability is enhanced obviously in the simulation as results.