Department of Applied Mathematics, National Chung Hsing University, Taichung City, 402, Taiwan.
Interdiscip Sci. 2018 Dec;10(4):823-835. doi: 10.1007/s12539-017-0254-3. Epub 2017 Jul 26.
The identification of genetic regulatory networks (GRNs) provides insights into complex cellular processes. A class of recurrent neural networks (RNNs) captures the dynamics of GRN. Algorithms combining the RNN and machine learning schemes were proposed to reconstruct small-scale GRNs using gene expression time series.
We present new GRN reconstruction methods with neural networks. The RNN is extended to a class of recurrent multilayer perceptrons (RMLPs) with latent nodes. Our methods contain two steps: the edge rank assignment step and the network construction step. The former assigns ranks to all possible edges by a recursive procedure based on the estimated weights of wires of RNN/RMLP (RE/RE), and the latter constructs a network consisting of top-ranked edges under which the optimized RNN simulates the gene expression time series. The particle swarm optimization (PSO) is applied to optimize the parameters of RNNs and RMLPs in a two-step algorithm. The proposed RE-RNN and RE-RNN algorithms are tested on synthetic and experimental gene expression time series of small GRNs of about 10 genes. The experimental time series are from the studies of yeast cell cycle regulated genes and E. coli DNA repair genes.
The unstable estimation of RNN using experimental time series having limited data points can lead to fairly arbitrary predicted GRNs. Our methods incorporate RNN and RMLP into a two-step structure learning procedure. Results show that the RE using the RMLP with a suitable number of latent nodes to reduce the parameter dimension often result in more accurate edge ranks than the RE using the regularized RNN on short simulated time series. Combining by a weighted majority voting rule the networks derived by the RE-RNN using different numbers of latent nodes in step one to infer the GRN, the method performs consistently and outperforms published algorithms for GRN reconstruction on most benchmark time series. The framework of two-step algorithms can potentially incorporate with different nonlinear differential equation models to reconstruct the GRN.
遗传调控网络（GRN）的识别为复杂的细胞过程提供了深入的了解。一类递归神经网络（RNN）捕获了 GRN 的动态。结合 RNN 和机器学习方案的算法被提出，用于使用基因表达时间序列重建小规模 GRN。
我们提出了具有神经网络的新 GRN 重建方法。RNN 扩展到具有潜在节点的一类递归多层感知器（RMLP）。我们的方法包含两个步骤：边缘等级分配步骤和网络构建步骤。前者通过基于 RNN/RMLP（RE/RE）的估计权重的递归过程为所有可能的边缘分配等级，后者在最优 RNN 模拟基因表达时间序列的情况下，由排名最高的边缘构建网络。粒子群优化（PSO）应用于两步算法中 RNN 和 RMLP 的参数优化。在大约 10 个基因的小规模 GRN 的合成和实验基因表达时间序列上测试了提出的 RE-RNN 和 RE-RNN 算法。实验时间序列来自酵母细胞周期调控基因和大肠杆菌 DNA 修复基因的研究。
使用具有有限数据点的实验时间序列对 RNN 进行不稳定估计可能会导致相当任意的预测 GRN。我们的方法将 RNN 和 RMLP 纳入两步结构学习过程。结果表明，在短模拟时间序列上，使用具有适当数量潜在节点的 RMLP 的 RE 通常比使用正则化 RNN 的 RE 更能准确地估计边缘等级。通过在第一步中使用不同数量的潜在节点的 RE-RNN 来组合加权多数投票规则，推断出 GRN 的网络，该方法在大多数基准时间序列上的 GRN 重建表现一致，优于已发表的算法。两步算法的框架可以潜在地与不同的非线性微分方程模型结合，以重建 GRN。
