Benchmarking algorithms for gene regulatory network inference from single-cell transcriptomic data
Benchmarking algorithms for gene regulatory network inference from single-cell transcriptomic data
复制标题
DOI：
10.1038/s41592-019-0690-6
复制
发表时间：
2020-01-06
期刊：
影响因子：
48
通讯作者：
Murali, T. M.
中科院分区：
文献类型：
--
作者：
Pratapa, Aditya;Jalihal, Amogh P.;Murali, T. M.
我们提出了一个系统的评估国家的最先进的算法推断基因调控网络从单细胞转录数据。作为评估准确性的基础事实，我们使用具有可预测轨迹的合成网络，文献策划的布尔模型和多种转录调控网络。我们开发了一种策略来模拟合成和布尔网络的单细胞转录数据，避免了以前使用的方法的陷阱。此外，我们从多个实验性单细胞RNA-seq数据集收集网络。我们开发了一个名为BEELINE的评估框架。我们发现这些算法的查准率-查全率曲线下面积和早期查准率都是中等的。该方法在恢复合成网络中的交互比布尔模型更好。对于布尔模型，具有最佳早期精度值的算法在实验数据集上也表现良好。不需要伪时间排序单元的技术通常更准确。根据这些结果，我们向最终用户提出了建议。BEELINE将帮助开发基因调控网络推理算法。使用合成和实验单细胞RNA-seq数据集对基因调控网络推理算法进行全面评估，发现异质性性能并向用户提出建议。
We present a systematic evaluation of state-of-the-art algorithms for inferring gene regulatory networks from single-cell transcriptional data. As the ground truth for assessing accuracy, we use synthetic networks with predictable trajectories, literature-curated Boolean models and diverse transcriptional regulatory networks. We develop a strategy to simulate single-cell transcriptional data from synthetic and Boolean networks that avoids pitfalls of previously used methods. Furthermore, we collect networks from multiple experimental single-cell RNA-seq datasets. We develop an evaluation framework called BEELINE. We find that the area under the precision-recall curve and early precision of the algorithms are moderate. The methods are better in recovering interactions in synthetic networks than Boolean models. The algorithms with the best early precision values for Boolean models also perform well on experimental datasets. Techniques that do not require pseudotime-ordered cells are generally more accurate. Based on these results, we present recommendations to end users. BEELINE will aid the development of gene regulatory network inference algorithms.Comprehensive evaluation of algorithms for inferring gene regulatory networks using synthetic and experimental single-cell RNA-seq datasets finds heterogeneous performance and suggests recommendations to users.