机构地区: 广东工业大学计算机学院
出 处: 《计算机应用研究》 2008年第1期50-52,共3页
摘 要: T细胞表位预测技术对于减少实验合成重叠肽,理解T细胞介导的免疫特异性和研制亚单位多肽及基因疫苗均有重要意义。为弥补已有基于机器学习方法的T细胞表位预测模型的可理解性的不足并进一步提高模型的预测精度,首先通过肽的预处理构建出了存储等长肽段的决策表,而后提出了基于粗糙集的分类器集成算法。该算法不但综合利用了基于信息熵的属性约简完备算法和其他属性约简算法的优势,而且将T细胞表位预测领域中的锚点知识融入到了属性值约简过程中。最后利用该算法来预测MHCⅡ类分子HLA-DR4(B1*0401)的结合肽,首次提取出了预测精度高且能帮助专家理解MHC分子与抗原肽的结合机理的产生式规则,为下一步的分子建模工作奠定了基础。 Predicting which peptides can bind to a specific MHC molecule is indispensable to minimizing the number of peptides required to synthesize, to the development of vaccines, and especially to aiding to understand the specificity of T-cell mediated immunity. In order to make up for the disadvantage of the existing T cell epitope prediction methods based on machine learning in understandability, a decision table comprising the nonamers was constructed by peptide preprocessing, then the multi-classifier integration algorithm based on rough sets was proposed, which took advantage of expert knowledge of binding motifs and diverse attribute reduction algorithms. Finally, with the help of the RSEN, the comprehensible rule set ensemble with strong generalization ability to predict the peptides that bind to HLA-DR4( B1 * 0401 ) was acquired.
领 域: [自动化与计算机技术] [自动化与计算机技术]