蛋白质超家族模体保守性及物理化学性质的分析

分析了全β类4个典型的蛋白质超家族中模体的功能,发现免疫球蛋白超家族和纤维结合蛋白类型Ⅲ超家族中的模体有相似的结构,但是它们行使不同的功能.血小板-白细胞C激酶底物的同源物结构域超家族和核酸结合超家族中的模体类型较多,虽然这些模体只是部分结构相似,然而它们却在各自的超家族中分别执行着相同的功能.文章进一步运用统计学方法研究了蛋白质超家族中保守模体的亲疏水特征、物理化学特征和结构特征.结果表明,模体差异有显著意义的残基存在于序列模体的保守位点上,相同的序列模体具有相似的二级结构.这些特征将对进一步识别超家族提供帮助.     

Reevaluation of hydropathy profiles of voltage-gated ionic channels

A reevaluation of the secondary structure of Na, Ca and K channel proteins led to the following results. Only three segments (S1, S5 and S6) of each repeat are sufficiently hydrophobic to be predicted as transmembrane helices, if a window of 19 amino acids is used. Some of the S2 and S3 segments show higher hydrophobic values when calculated with the window of 9 amino acids and can be predicted as short helices. S4 segments are strongly hydrophilic and cannot be predicted as transmembrane helices. Some of the S2, S3 and S4 segments have an amphipathic character; however, these helices do not span a membrane. A model is proposed where 12 hydrophobic transmembrane helices surround 12 shorter helices, forming a hydrophilic pore. In addition, a unique pattern for S4 segments of voltage-gated channel proteins is defined.     

基于氨基酸亲疏水分布的最小离散增量方法识别蛋白质超家族

摆脱目前常用的基于特征模体来识别超家族的方法,引入单个氨基酸的物理化学性质(亲疏水性质)及其在氨基酸序列中的分布特征参数(分段的方法),利用最小离散增量算法对属于同一结构类的不同超家族进行了识别,总预测成功率令人满意.选取了全α类、全β类、α β类、α/β类中的各4个超家族分别进行了识别.对全α类的4个超家族,se lf-cons istency检验和jack-kn ife检验可达83.0%,81.2%;对全β类的4个超家族,两种检验均为80.9%;对α β类的4个超家族,两种检验总成功率最高分别为88.6%,88.0%;识别α/β类的4个超家族时两种检验的结果分别为69.3%和67.6%.