An RNA Scoring Function for Tertiary Structure Prediction Based on Multi-Layer Neural Networks
An RNA Scoring Function for Tertiary Structure Prediction Based on Multi-Layer Neural Networks
复制标题
基于多层神经网络的三级结构预测RNA评分函数
DOI：
10.1134/s0026893319010175
复制
发表时间：
2019
影响因子：
1.2
通讯作者：
J. Zhang
中科院分区：
文献类型：
--
作者：
Y. Z. Wang;J. Zhang
良好的评分函数对于 RNA 三级结构的从头预测是必要的。在这项研究中，我们探索了基于机器学习的方法作为评分函数的强大功能。与传统的评分函数相比，该方法在融合不同类型的特征方面更加灵活；它也没有选择参考状态的难题。构建并训练了两个多层神经网络。他们将 RNA 结构候选作为输入，然后输出其相似度分数，以评估候选结构与天然结构的相似度。第一个网络在 RNA 结构的粗粒度水平上工作，而第二个网络在全原子水平上工作。我们还构建了一个 RNA 数据库，并将其分为训练集、验证集和测试集，分别包含 322、70 和 70 个 RNA。每个 RNA 都附有 300 个通过高温分子动力学模拟生成的诱饵。网络在训练集上进行训练，然后根据验证集的损失使用提前停止策略进行优化。然后我们在测试集上测试了网络的性能。结果发现始终优于最近基于知识的全原子势。
A good scoring function is necessary for ab inito prediction of RNA tertiary structures. In this study, we explored the power of a machine learning based approach as a scoring function. Compared with the traditional scoring functions, the present approach is more flexible in incorporating different kinds of features; it is also free of the difficult problem of choosing the reference state. Two multi-layer neural networks were constructed and trained. They took RNA a structural candidate as input and then output its likeness score that evaluates the likeness of the candidate to the native structure. The first network was working at the coarse-grained level of RNA structures, while the second at the all-atom level. We also built an RNA database and split it into the training, validation, and testing sets, containing 322, 70, and 70 RNAs, respectively. Each RNA was accompanied with 300 decoys generated by high-temperature molecular dynamics simulations. The networks were trained on the training set and then optimized with an early-stop strategy, based on the loss of the validation set. We then tested the performance of the networks on the testing set. The results were found to be consistently better than a recent knowledge-based all-atom potential.