Voigt Christopher A, Martinez Carlos, Wang Zhen-Gang, Mayo Stephen L, Arnold Frances H
Biochemistry and Molecular Biophysics, California Institute of Technology, mail code 210-41, Pasadena, California 91125, USA.
Nat Struct Biol. 2002 Jul;9(7):553-8. doi: 10.1038/nsb805.
Borrowing concepts from the schema theory of genetic algorithms, we have developed a computational algorithm to identify the fragments of proteins, or schemas, that can be recombined without disturbing the integrity of the three-dimensional structure. When recombination leaves these schemas undisturbed, the hybrid proteins are more likely to be folded and functional. Crossovers found by screening libraries of several randomly shuffled proteins for functional hybrids strongly correlate with those predicted by this approach. Experimental results from the construction of hybrids of two beta-lactamases that share 40% amino acid identity demonstrate a threshold in the amount of schema disruption that the hybrid protein can tolerate. To the extent that introns function to promote recombination within proteins, natural selection would serve to bias their locations to schema boundaries.
借鉴遗传算法的模式理论概念,我们开发了一种计算算法,用于识别蛋白质片段或模式,这些片段或模式在不干扰三维结构完整性的情况下可以进行重组。当重组不干扰这些模式时,杂合蛋白更有可能折叠并具有功能。通过筛选几个随机改组的蛋白质文库以寻找功能性杂合体而发现的交叉与该方法预测的交叉密切相关。构建具有40%氨基酸同一性的两种β-内酰胺酶杂合体的实验结果表明,杂合蛋白能够耐受的模式破坏量存在一个阈值。就内含子在促进蛋白质内重组方面的作用而言,自然选择将倾向于使它们的位置偏向模式边界。