Kolaskar A S, Ramabrahmam V, Soman K V
Int J Pept Protein Res. 1980 Jul;16(1):1-11. doi: 10.1111/j.1399-3011.1980.tb02929.x.
A simple algorithm has been developed to detect beta-bends and 'loops'-chain reversals containing five amino acid residues, using only coordinates of C alpha-atoms from crystal structure data of globular proteins using the above algorithm. Analysis of bends have showed that the total number of bends in each protein (TB) is linearly related to total number of non-hydrophobic residues in that protein which in turn is related linearly to total number of amino acid residues. Secondly, we found that a large number of consecutive bends occur in each protein which give rise to on an average only three independent residues per turn. Positional preference of amino acid residues in chain reversals is stressed. Consideration of pairs of amino acid residues in positions (i + 1) and (i + 2) of bends seems to provide a more reliable basis for predicting chain reversals in proteins.
已经开发出一种简单算法,仅使用球状蛋白质晶体结构数据中Cα原子的坐标来检测β-转角和含有五个氨基酸残基的“环”-链反向结构。使用上述算法对转角的分析表明,每种蛋白质中转角的总数(TB)与该蛋白质中非疏水残基的总数呈线性相关,而非疏水残基的总数又与氨基酸残基的总数呈线性相关。其次,我们发现每种蛋白质中都有大量连续的转角,平均每转仅产生三个独立残基。强调了链反向结构中氨基酸残基的位置偏好。考虑转角位置(i + 1)和(i + 2)处的氨基酸残基对似乎为预测蛋白质中的链反向结构提供了更可靠的基础。