Tissue-specificity of cancer: linking the rewiring of molecular networks in different tissues and the mutational profiles of different cancer types

遗传性疾病通常只影响人体中的一小部分组织，即使相关基因普遍表达。这些疾病中的许多损害了信号通路的功能。例如，癌症会干扰细胞生长和凋亡相关信号通路的活性。这些信号通路不是静态的，而是根据组织显示不同的布线和激活模式。在这里，我们将研究组织特异性信号是如何影响细胞对疾病的选择性易感性的。信号通路由严格控制的蛋白质结合事件级联组成。研究信号通路的组织特异性功能的先决条件是了解哪些蛋白质-蛋白质相互作用（PPI）仅在所有组织的一个子集中实现。由于选择性剪接和蛋白质丰度已被证明以组织特异性方式重塑PPI网络，我们将使用来自不同人体组织的RNA测序和蛋白质组学数据来预测哪些先前测量的PPI在哪些组织中实现。我们将使用质谱法在不同细胞类型中确定差异蛋白质复合物的形成，以验证我们的方法预测组织特异性PPI网络的能力。我们还将利用遗传变异数据的组织特异性分布，假设仅在源自某些组织的癌症中发现的突变在具有组织的基因组中富集，具体功能。通过解决识别子网络的优化问题，这些子网络在生物学上合理的约束下最好地解释了癌症突变的组织特异性分布，我们将同时对组织特异性通路布线进行逆向工程，并为细胞对疾病突变的组织特异性脆弱性提供机制解释。我们将通过展示组织特异性通路组分如何以组织特异性方式调节通路活性并触发癌症相关表型来实验验证我们的发现。

Genetic diseases usually affect only a small subset of tissues in the human body even when the responsible genes are expressed ubiquitously. Many of these diseases impair the function of signaling pathways. Cancer, for example, interferes with the activity of cell growth- and apoptosis-related signaling pathways. These signaling pathways are not static but show different wiring and activation patterns depending on the tissue. Here, we will investigate how tissue-specific signaling contributes to the selective vulnerability of cells to disease.Signaling pathways are composed of tightly controlled cascades of protein binding events. A prerequisite to study the tissue-specific function of signaling pathways is the knowledge about which protein-protein interactions (PPIs) are realized only in a subset of all tissues. As alternative splicing and protein abundances have been shown to remodel PPI networks in a tissue-specific manner, we will use RNA sequencing and proteomics data from different human tissue to predict which previously measured PPIs are realized in which tissues. We will determine differential protein complex formation using mass spectrometry in different cell types to validate the capability of our approach to predict tissue-specific PPI networks.To use tissue-specific PPIs to predict events of tissue-specific signaling, we will additionally leverage the tissue-specific distribution of genetic variation data assuming that mutations only found in cancers originating from certain tissues are enriched in gene sets with tissue-specific functions. By solving the optimization problem of identifying subnetworks that explain the tissue-specific distribution of cancer mutations under biologically plausible constraints best, we will simultaneously reverse engineer tissue-specific pathway wiring and provide mechanistic explanations for the tissue-specific vulnerability of cells to disease mutations. We will experimentally validate our findings by demonstrating how tissue-specific pathway components can tune pathway activity and trigger cancer-related phenotypes in a tissue-specific manner.