Bogardt R A, Jones B N, Dwulet F E, Garner W H, Lehman L D, Gurd F R
J Mol Evol. 1980 Jul;15(3):197-218. doi: 10.1007/BF01732948.
Multivariate statistical analyses were applied to 16 physical and chemical properties of amino acids. Four of these properties; volume, polarity, isoelectric point (charge), and hydrophobicity were found to explain adequately 96% of the total variance of amino acid attributes. Using these four quantitative measures of amino acid properties, a structural discriminate function in the form of a weighted difference sum of squares equation was developed. The discriminate function is weighted by the location of each particular residue within a given tertiary structure and yields a numerical discriminate or difference value for the replacement of these residues by different amino acids. This resulting discriminate value represents an expression of the perturbation in the local positional environment of a protein when an amino acid substitution occurs. With the use of this structural discriminate function, a residue by residue comparison of the known mammalian myoglobin sequences was carried out in an attempt to elucidate the positions of possible deviations from the known tertiary structure of sperm whale myoglobin. Only 11 of the 153 residue positions in myoglobin demonstrated possible structural deviations. From this analysis, indices of difference were calculated for all amino acid exchanges between the various myoglobins. All comparisons yielded indices of difference that were considerably lower than would be expected if mutations had been fixed at random, even if the organization of the genetic code is taken into consideration. On the basis of these results, it is inferred that some form of selection has acted in the evolution of mammalian myoglobins to favor amino acid substitutions that are compatible with the retention of the original conformation of the protein.
多变量统计分析应用于氨基酸的16种物理和化学性质。其中四种性质:体积、极性、等电点(电荷)和疏水性,被发现足以解释氨基酸属性总方差的96%。利用这四种氨基酸性质的定量测量方法,开发了一种以加权差平方和方程形式的结构判别函数。该判别函数由给定三级结构中每个特定残基的位置加权,并为用不同氨基酸替换这些残基产生一个数值判别或差值。这个产生的判别值代表了氨基酸替换发生时蛋白质局部位置环境中扰动的一种表达形式。通过使用这种结构判别函数,对已知的哺乳动物肌红蛋白序列进行逐个残基比较,试图阐明与抹香鲸肌红蛋白已知三级结构可能存在偏差的位置。肌红蛋白153个残基位置中只有11个显示出可能的结构偏差。通过该分析,计算了各种肌红蛋白之间所有氨基酸交换的差异指数。所有比较得出的差异指数都大大低于如果突变是随机固定的情况下预期的值,即使考虑到遗传密码的组织形式也是如此。基于这些结果,可以推断在哺乳动物肌红蛋白的进化过程中,某种形式的选择作用有利于与蛋白质原始构象保留相兼容的氨基酸替换。