MOE Key Laboratory of Bioinformatics, School of Life Sciences, Tsinghua University, Beijing, 100084, China.
Beijing Advanced Innovation Center for Structural Biology, Tsinghua University, Beijing, 100084, China.
BMC Bioinformatics. 2020 Apr 3;21(1):133. doi: 10.1186/s12859-020-3476-z.
Despite the great advance of protein structure prediction, accurate prediction of the structures of mainly β proteins is still highly challenging, but could be assisted by the knowledge of residue-residue pairing in β strands. Previously, we proposed a ridge-detection-based algorithm RDbC that adopted a multi-stage random forest framework to predict the β-β pairing given the amino acid sequence of a protein.
In this work, we developed a second version of this algorithm, RDbC2, by employing the residual neural network to further enhance the prediction accuracy. In the benchmark test, this new algorithm improves the F1-score by > 10 percentage points, reaching impressively high values of ~ 72% and ~ 73% in the BetaSheet916 and BetaSheet1452 sets, respectively.
Our new method promotes the prediction accuracy of β-β pairing to a new level and the prediction results could better assist the structure modeling of mainly β proteins. We prepared an online server of RDbC2 at http://structpred.life.tsinghua.edu.cn/rdb2c2.html.
尽管蛋白质结构预测取得了巨大进展，但准确预测主要为β的蛋白质的结构仍然极具挑战性，但可以通过β链中残基-残基配对的知识来辅助。此前，我们提出了一种基于脊检测的算法 RDbC，该算法采用多阶段随机森林框架，根据蛋白质的氨基酸序列预测β-β配对。
在这项工作中，我们通过使用残差神经网络开发了该算法的第二个版本 RDbC2，以进一步提高预测准确性。在基准测试中，该新算法将 F1 得分提高了>10 个百分点，在 BetaSheet916 和 BetaSheet1452 数据集上分别达到了令人印象深刻的约 72%和 73%的高值。
我们的新方法将β-β 配对的预测准确性提升到了一个新的水平，预测结果可以更好地辅助主要为β的蛋白质的结构建模。我们在 http://structpred.life.tsinghua.edu.cn/rdb2c2.html 上准备了 RDbC2 的在线服务器。
