生物分子网络的统计构建与分析及其在油菜组学数据挖掘中的应用

Systems biology is one of the major strategic needs of our country’s basic research fields in the national medium and long-term science and technology planning, and yet it is also a prior to the deployment of development area and energetically promoted interdisciplinary field in the CSC’s twelfth and thirteenth five-year development plans. Bio-molecular networks and omics data mining are research focus in systems biology. Construction of bio-molecular networks based on data with few sample points, designing new topological indexes for topological analysis of bio-molecular networks, as well as identification of crucial bio-molecules are important issues in the fields of biological networks and omic data mining. Based on the above issues, this project will firstly perform theoretical analysis, including exploring the statistical construction of biological networks on the basis of data with few sample points, designing new topological measures by considering features of biological networks, and building new algorithms to identify crucial bio-molecules via extending the hidden Markov random field model and some other models. Afterwards, based on omics data for Brassica napus under drought, salt and other treatments, this project will explore applications of the above theoretical investigations. Related biological networks for Brassica napus will be constructed, stress-responsive crucial genes or transcriptional factors will be identified, molecular mechanisms of stress-responsiveness will be clarified, and PCR biological validations will be performed to verify the reproducibility of the identified genes. This project aims to design several novel algorithms on biological network construction and analysis, and then applies them to omics data mining of Brassica napus to explore stress-responsive crucial genes. The related investigations have great implications in the areas of biological network and omics data analysis, and it will also establish foundation to breed new crop varieties with improved stress resistance for molecular breeding researchers.

系统生物是国家中长期科技规划中面向国家重大战略需求的基础研究领域,是基金委十二、十三五发展规划优先部署和积极推动的交叉学科。生物分子网络及组学数据挖掘是系统生物的研究热点。少样本点数据的网络重构、网络拓扑结构新指标设计、关键分子识别等是生物网络及组学数据挖掘中的重要问题。本项目首先采用灰色关联分析等方法探讨少样本点数据生物网络的重构,结合生物网络的特征设计拓扑结构分析的新指标，推广隐马尔科夫随机场等模型设计关键分子识别算法,展开理论研究。然后基于干旱、盐碱等多种胁迫处理下的油菜组学数据开展应用研究。构建油菜相关生物网络并挖掘胁迫响应关键基因或转录因子,阐明胁迫响应的分子机制,并对识别的基因进行PCR生物学验证。本项目旨在设计若干新的生物网络统计构建与分析方法,并应用于油菜组学数据挖掘关键基因等。相关研究对生物网络及组学数据分析具有重要意义,为分子育种研究者培育具有抗性的作物新品种奠定基础。

网络构建、拓扑结构分析、关键节点识别是生物网络及组学数据挖掘的重要问题。本项目在单层生物网络的构建、分析及油菜组学数据分析方面取得如下成果：1）系统总结了生物网络的构建、建模与动力学分析、结构分析和关键信息挖掘、组学数据挖掘中的应用等进展；基于灰色关联分析，提出了针对转录组数据构建基因共表达和差异共表达网络的一般框架；提出基于变分Bayes方法重构加权网络的一般框架。2）提出识别复杂网络中的关键节点的若干算法，如SpectralRank，适用于二分图的SVD-rank 和SVDA-rank等；3）从全基因组层面识别出油菜五个转录因子家族的2167名成员, 通过多胁迫转录组数据分析挖掘出315个关键的转录因子, 利用统计和网络分析等揭示了油菜胁迫响应一些转录组基础，并在转基因油菜研究方面进行了初步探索；4）基于非结构化的新冠肺炎病例数据，通过统计分析和网络分析揭示了河南省新冠肺炎的流行病学特征和传播的阶段性特征及聚集爆发现象；5）研究了拟南芥和酵母菌的基因调控网络的结构可控性，阐明了一些功能基因在网络控制中的作用。相关研究已出版1部Springer专著，发表SCI/EI论文18篇，获河南省科技进步奖1项等。本项目研究不仅会促进网络科学的发展，而且将为作物遗传育种奠定若干理论基础。