整合多组学数据识别乳腺癌自噬相关非编码RNA参与的风险级联通路

Breast cancer is a serious malignant tumor that threatens the health of women, and its morbidity and mortality are increasing year by year. Recent studies have suggested that the process of autophagy plays an important role in the development and prognosis of breast cancer. In addition, non-coding RNAs are demonstrated to play important roles in the regulation of autophagy. Identification of autophagy associated risk cascades in breast cancer involved in non-coding RNAs contributes greatly to the study of breast cancer pathogenesis and prognosis. In this project, we first integrate transcriptional and post-transcriptional cascade regulations, combine with breast cancer high-throughput transcriptome data and build breast cancer specific cascade regulatory network. Next, combining with autophagy associated regulators, the breadth-first algorithm is used to identify the autophagy associated cascades. We design node scoring measure based on the cascade node mutation data and location information, calculate the area under the enrichment curve of differentially expressed genes in breast cancer and identify risk autophagy-associated cascades. Based on the identified risk cascades, we will evaluate the effects of autophagy for the pathogenesis and survival prognosis of breast cancer, and predict potential drugs for the treatment of breast cancer. Finally, we integrate the developed methods and results of this project to build a disease risk cascade annotation platform, providing data query, visualization and online analysis. The project possesses important scientific significance and application value for studying the pathogenesis and prognosis of breast cancer affected by autophagy, and provides an important tool for the analysis of the relationship between regulatory cascades involved in non-coding RNAs and complex diseases.

乳腺癌是严重威胁女性健康的恶性肿瘤，其发病率和死亡率逐年上升。研究表明自噬在乳腺癌的发生发展和预后治疗中扮演重要角色。此外，非编码RNA在自噬过程的调控中发挥重要作用。识别非编码RNA参与的乳腺癌自噬相关风险级联通路对研究乳腺癌发病机制和预后治疗具有重要意义。本项目整合转录及转录后调控因子的调控关系，结合乳腺癌高通量转录组数据，构建非编码RNA参与的乳腺癌级联调控网络；采用广度优先算法，结合自噬相关因子，识别自噬相关级联通路；根据通路节点突变数据和位置信息，设计节点打分测度，计算乳腺癌差异表达基因的通路富集曲线下面积得分，识别自噬相关风险级联通路；分析自噬影响的乳腺癌发病机制和生存预后，为乳腺癌的治疗提供潜在药物。最后，整合本项目研究成果，构建疾病风险通路查询分析平台。本项目对研究自噬影响的乳腺癌发病机制和预后治疗具有科学意义和应用价值，为研究非编码RNA参与的疾病风险通路提供重要工具。

本研究项目按照规定计划如期完成。乳腺癌是严重威胁女性健康的恶性肿瘤，其发病率和死亡率逐年上升。研究表明自噬在乳腺癌的发生发展和预后治疗中扮演重要角色。识别非编码RNA参与的乳腺癌自噬相关风险级联通路对研究乳腺癌发病机制和预后治疗具有重要意义。本项目整合基因转录及转录后调控因子的级联调控关系，结合乳腺癌高通量转录组数据，构建非编码RNA参与的乳腺癌级联调控网络，387个转录因子（TF），174个microRNA（miRNA），407个长非编码 RNA（lncRNA）和905个PCG。然后采用随机游走算法，结合已知的自噬相关因子，识别乳腺癌中自噬相关的非编码RNA因子，包括26个miRNAs和19个lncRNAs。进一步结合乳腺癌致病基因，基于乳腺癌特异转录调控网络，识别了乳腺癌相关的疾病风险因子，包括lncRNA、circRNA和eRNA等。本项目进而开发机器学习模型，融合肿瘤基因组和表观基因组学特征，分析肿瘤耐药性的产生机制和改进方法，通过芯片探针重注释的方法，采用KS统计检验，为恶性肿瘤的治疗提供候选小分子药物。在成果转化方面，本项目构建了基于非编码RNA的复杂疾病风险因子查询和疾病治疗小分子药物优选分析平台，为研究非编码RNA在疾病进展和治疗中的作用提供重要帮助。在本课题的资助下，共发表SCI论文6篇（均对本项目号进行了标注），主要发表于国外著名生命科学相关杂志《BMC genomics》、《RNA biology》及《Frontiers in Bioengineering and Biotechnology》等，累计影响因子达到29.302。