Signaling pathways are responsible for the regulation of cell processes, such as monitoring the external environment, transmitting information across membranes, and making cell fate decisions. Given the increasing amount of biological data available and the recent discoveries showing that many diseases are related to the disruption of cellular signal transduction cascades, in silico discovery of signaling pathways in cell biology has become an active research topic in past years. However, reconstruction of signaling pathways remains a challenge mainly because of the need for systematic approaches for predicting causal relationships, like edge direction and activation/inhibition among interacting proteins in the signal flow. We propose an approach for predicting signaling pathways that integrates protein interactions, gene expression, phenotypes, and protein complex information. Our method first finds candidate pathways using a directed-edge-based algorithm and then defines a graph model to include causal activation relationships among proteins, in candidate pathways using cell cycle gene expression and phenotypes to infer consistent pathways in yeast. Then, we incorporate protein complex coverage information for deciding on the final predicted signaling pathways. We show that our approach improves the predictive results of the state of the art using different ranking metrics.

中文翻译：

---

发现信号通路中的细胞周期和蛋白质复合物动力学

信号通路负责调节细胞过程，例如监测外部环境、跨膜传递信息以及做出细胞命运决定。鉴于可用的生物学数据越来越多，并且最近发现许多疾病与细胞信号转导级联的破坏有关，因此在细胞生物学中通过计算机发现信号通路已成为过去几年的活跃研究课题。然而，信号通路的重建仍然是一个挑战，主要是因为需要系统的方法来预测因果关系，如边缘方向和信号流中相互作用蛋白质之间的激活/抑制。我们提出了一种预测信号通路的方法，该方法整合了蛋白质相互作用、基因表达、表型、和蛋白质复合物信息。我们的方法首先使用基于有向边的算法找到候选途径，然后定义一个图形模型以包括蛋白质之间的因果激活关系，在候选途径中使用细胞周期基因表达和表型来推断酵母中的一致途径。然后，我们结合蛋白质复合物覆盖信息来决定最终预测的信号通路。我们表明，我们的方法使用不同的排名指标改进了现有技术的预测结果。我们结合蛋白质复合物覆盖信息来决定最终预测的信号通路。我们表明，我们的方法使用不同的排名指标改进了现有技术的预测结果。我们结合蛋白质复合物覆盖信息来决定最终预测的信号通路。我们表明，我们的方法使用不同的排名指标改进了现有技术的预测结果。