Department of Molecular Biology, University of Wyoming, Laramie, WY 82071, USA.
J Mol Evol. 2011 Aug;73(1-2):23-33. doi: 10.1007/s00239-011-9454-3. Epub 2011 Jul 29.
For high-throughput structural genomic and evolutionary bioinformatics approaches, there is a clear need for fast methods to evaluate substitutions structurally. Coarse-grained methods are both powerful and fast, and a coarse-grained approach to position the substituted side chains is presented. Through the application of a coarse-grained method, a speed-up on the single- residue replacement, of at least sevenfold is achieved compared with modern all-atom approaches. At the same time, this approach maintains a small median RMSD from the leading all-atom approach (as measured in coarse-grained space), and predicts the conformation of point mutants with similar accuracy and generates biologically realistic side chain angles. This method is also substantially more predictable in its run time, making it useful for high-throughput studies of protein structural evolution. To demonstrate the utility of this method, it has been implemented in a forward simulation of sequences threaded through the SH2 domains, with selective pressures to fold and bind specifically. The relative substitution rates across the protein structure and at the binding interface are reflective of those observed in SH2 domain evolution. The algorithm has been implemented in C++, with the source code and binaries (currently supported for Linux systems) freely available as SARA at http://www.wyomingbioinformatics.org/LiberlesGroup/SARA .
对于高通量结构基因组学和进化生物信息学方法，显然需要快速的方法来从结构上评估取代。粗粒化方法既强大又快速，本文提出了一种粗粒化方法来定位取代的侧链。通过应用粗粒化方法，与现代全原子方法相比，单残基替换的速度至少提高了 7 倍。同时，与领先的全原子方法相比（在粗粒化空间中测量），该方法保持较小的中值 RMSD，并且以相似的精度预测点突变体的构象，并生成具有生物学合理性的侧链角度。该方法的运行时间也更具可预测性，因此非常适合高通量蛋白质结构进化研究。为了证明该方法的实用性，已经在通过 SH2 结构域进行序列的正向模拟中实现了该方法，这些模拟具有折叠和特异性结合的选择性压力。整个蛋白质结构和结合界面上的相对取代率反映了在 SH2 结构域进化中观察到的取代率。该算法已用 C++实现，源代码和二进制文件（目前支持 Linux 系统）可在 http://www.wyomingbioinformatics.org/LiberlesGroup/SARA 上免费获得。
