Gates R E, Fisher H F
Proc Natl Acad Sci U S A. 1971 Dec;68(12):2928-31. doi: 10.1073/pnas.68.12.2928.
A model for globular protein molecules based on a linear and random sequence of polar and nonpolar amino acids was developed. Polar amino acids were assumed to be always in contact with the aqueous solvent, while nonpolar amino acids were assumed to have a tendency to be buried. Considerations of amino-acid dimensions indicated that only runs of four or more nonpolar amino acids in a row could allow some amino acids to be more than 1-nm (10-A) removed from the surface of the protein. The expected volume fraction of a protein molecule that could be more than 1-nm removed from the surface was obtained with the assumption of a random sequence. Calculation of this volume fraction for a number of simple geometric shapes indicated that some nonpolar amino acids must be exposed to solvent and that the maximum average thickness of globular proteins should be 3-4 nm. Good agreement with published protein dimensions was obtained.
基于极性和非极性氨基酸的线性随机序列，建立了一种球状蛋白质分子模型。假定极性氨基酸始终与水性溶剂接触，而非极性氨基酸则倾向于被埋藏。对氨基酸尺寸的考量表明，只有连续四个或更多非极性氨基酸的排列，才会使某些氨基酸与蛋白质表面的距离超过1纳米（10埃）。在随机序列的假设下，得出了蛋白质分子中可能与表面距离超过1纳米的预期体积分数。对一些简单几何形状的该体积分数进行计算表明，一些非极性氨基酸必须暴露于溶剂中，且球状蛋白质的最大平均厚度应为3至4纳米。这与已发表的蛋白质尺寸数据吻合良好。
