摘要
基于氨基酸一级序列,应用离散增量的方法识别ArchDB数据库中的6 100个β发夹模体和2 701个非β发夹模体。模体按照二级结构中无规卷曲的长进行分类得到"07232"、"07322"、"09432"、"09522"、"13643"、"13733"、"13832"、"13922"型。以氨基酸和氨基酸紧邻关联为参量,利用10-fold交叉检验的方法进行检验,平均识别精度均达到75.0%和83.0%以上。以氨基酸亲疏水性和亲疏水紧邻关联为参量进行检验,平均识别精度有所降低。
Based on the primary sequence of amino acids,6 100 β-hairpin motifs and 2 701 non-hairpins motifs of the ArchDB database were recognized with the method of increment of diversity.β3-hairpin motifs were classified by the coil length in the secondary structure,into “07232 ”,”07322”、“09432”、“09522”、“13643”、“13733”、“13832” and “13922”types.Using amino acid and amino acids tight neighbor correlation as parameters,β-hairpin motifs were tested by the 10-fold cross-validation method and the average recognition accuracy could reached 75.0％ and 83.0％.Using amino acid hydrophilic and hydrophobic,hydrophilic and hydrophobic tight neighbor correlation as the parameters,the average recognition accuracy decreased.
出处
《湖北农业科学》
北大核心
2013年第23期5898-5901,共4页
Hubei Agricultural Sciences
基金
辽宁省教育厅教学改革立项项目(2012411)
关键词
β发夹模体
离散增量
离散量
β-hairpin motif
increment of diversity
measure of diversity
