Performance assessment of sample-specific network control methods for bulk and single-cell biological data analysis.
Performance assessment of sample-specific network control methods for bulk and single-cell biological data analysis.
复制标题
用于批量和单细胞生物数据分析的样本特定网络控制方法的性能评估
DOI：
10.1371/journal.pcbi.1008962
复制
发表时间：
2021-05
影响因子：
4.3
通讯作者：
Zeng T
中科院分区：
文献类型：
--
作者：
Guo WF;Yu X;Shi QQ;Liang J;Zhang SW;Zeng T
在过去的几年中，已经提出了丰富的样本特定的网络构建方法和结构化网络控制方法来识别样本特定的驱动节点，以支持生物网络系统的样本特定的网络控制（SSC）分析。然而，目前还没有对这些最先进的方法进行综合评价。在这里，我们进行了性能评估的16个SSC分析工作流程，通过使用4个样本特定的网络重建方法和4个代表性的结构控制方法的组合。该研究包括对代表性生物网络的模拟评估，对多个癌症批量表达数据集的个性化驱动基因进行优先级排序，以及来自TCGA的匹配患者样本，以及单细胞RNA-seq数据集上与细胞分化相关的细胞标记基因和关键时间点识别。通过广泛比较现有SSC分析工作流程的分析，我们提供了以下建议和分组工作流程。(i)网络控制方法的性能强烈依赖于上游样本特定网络方法，并且小区特定网络构造（CSN）方法和单样本网络（SSN）方法是优选的样本特定网络构造方法。(ii)在构建样本特定网络后，基于无向网络的控制方法比基于有向网络的控制方法更有效。此外，这些数据和评估管道可在https://github.com/WilfongGuo/Benchmark_control上免费获得。最近，利用复杂网络的结构可控性的结构网络控制方法已经成为用于识别驱动节点（例如，驱动基因或药物靶点）。实际上，将结构网络控制应用于生物网络系统的关键是如何刻画这些生物系统的状态转移网络（或相互作用网络）。状态转移网络是一个图，其中节点表示系统变量，边表示触发系统从一个吸引子（例如，个体患者的健康状态）到另一吸引子（例如，个别患者的病情）。状态转移网络刻画了网络系统在任意两个吸引子之间的状态转移。无独有偶，最近提出了单样本网络构造方法来构造样本特异性相互作用网络，以表征样本特异性生物系统的状态转移。因此，现在是时候进行一项研究，以评估这些国家的最先进的方法在现实世界的生物系统，包括样本特定的网络构建方法和结构网络控制方法的性能。在这里，我们全面提供了16个分析工作流程的性能评估，这些工作流程基于四种样本特定的网络重建方法和四种代表性的结构网络控制方法在各种真实世界生物数据集上的组合。通过这项研究，我们揭示了不同的分析框架的相对行为和性能，以评估和推荐这些方法在特定的生物应用场景。此外，我们想指出一些挑战，目前的方法，并提出了几个发展方向，新的SSC分析工作流程时，面临更复杂的实践条件下的控制原则。
In the past few years, a wealth of sample-specific network construction methods and structural network control methods has been proposed to identify sample-specific driver nodes for supporting the Sample-Specific network Control (SSC) analysis of biological networked systems. However, there is no comprehensive evaluation for these state-of-the-art methods. Here, we conducted a performance assessment for 16 SSC analysis workflows by using the combination of 4 sample-specific network reconstruction methods and 4 representative structural control methods. This study includes simulation evaluation of representative biological networks, personalized driver genes prioritization on multiple cancer bulk expression datasets with matched patient samples from TCGA, and cell marker genes and key time point identification related to cell differentiation on single-cell RNA-seq datasets. By widely comparing analysis of existing SSC analysis workflows, we provided the following recommendations and banchmarking workflows. (i) The performance of a network control method is strongly dependent on the up-stream sample-specific network method, and Cell-Specific Network construction (CSN) method and Single-Sample Network (SSN) method are the preferred sample-specific network construction methods. (ii) After constructing the sample-specific networks, the undirected network-based control methods are more effective than the directed network-based control methods. In addition, these data and evaluation pipeline are freely available on https://github.com/WilfongGuo/Benchmark_control. Recently structural network control methods which exploit structural controllability of complex networks have become a powerful analysis tool for identifying driver nodes (e.g., driver genes or drug targets) in biological networked systems. In fact, the key point to apply structural network control on biological networked systems is how to characterize the state transition network (or interaction network) of these biological systems. The state transition network is a graph in which nodes denote the system variables and edges denote the significant interactions to trigger the state transition of system from one attractor (e.g., healthy state of individual patients) to another attractor (e.g., disease state of individual patients). The state transition network characterizes the state transition of networked system between any two attractors. Coincidentally, recently single sample network construction methods have been proposed to construct sample specific interaction network for characterizing the state transition of sample-specific biological systems. Therefore it is time to conduct a study to assess the performance of these state-of-the-art methods on the real-world biological systems, including sample-specific network construction methods and structural network control methods. Here we comprehensively provided a performance evaluation of 16 analysis workflows based on the combination of four sample-specific network reconstruction methods and four representative structural network control methods on a variety of real-world biological datasets. By this study, we have shed light on the relative behavior and performance of different analysis frameworks for evaluating and recommending these methods in particular biological application scenarios. In addition, we would like to point a few challenges for current methods and suggest several development directions of new SSC analysis workflows when more complex practice conditions of control principles are faced.
登录
查看更多内容
使用单样本动态网络生物标志物量化复杂疾病的关键状态
影响因子：
4.3
作者：
Liu X;Chang X;Liu R;Yu X;Chen L;Aihara K
通讯作者：
Aihara K
影响因子：
14.9
作者：
Kanehisa M;Goto S;Sato Y;Furumichi M;Tanabe M
通讯作者：
Tanabe M
一种寻找目标控制复杂网络最佳驱动节点的新算法及其在药物靶标识别中的应用
影响因子：
4.4
作者：
Guo WF;Zhang SW;Shi QQ;Zhang CM;Zeng T;Chen L
通讯作者：
Chen L
使用样本特定网络对疾病进行个性化表征
影响因子：
14.9
作者：
Liu X;Wang Y;Ji H;Aihara K;Chen L
通讯作者：
Chen L
通过动态网络生物标志物寻找乳腺癌内分泌抵抗过程的临界点
影响因子：
5.5
作者：
Liu R;Wang J;Ukai M;Sewon K;Chen P;Suzuki Y;Wang H;Aihara K;Okada-Hatakeyama M;Chen L
通讯作者：
Chen L