Statistical Methods for Estimation of Gene Regulatory Networks

描述(申请人提供)：基因组技术的进步导致了许多基因的发现，这些基因的表达在不同的细胞条件下是不同的；然而，基因并不是单独作用的，而是在驱动细胞功能的复杂网络中共同作用。通过考虑基因(和基因产物)之间的相互作用，人们对潜在的细胞机制有了更深入的了解。对这些基因调控网络的估计对于理解细胞机制、检测细胞类型之间的差异以及预测细胞对干预的反应是必要的。癌症进展已被证明会在对正常细胞功能至关重要的遗传网络中产生剧烈变化。一些致癌基因突变会在网络结构中产生自我维持的变化，因此移除原始突变并不能恢复正常的细胞功能。这表明，识别原始致癌突变可能不足以进行有针对性的干预；相反，可能需要详细了解正常和恶性细胞中存在的基因调控网络。基因扰动实验是研究基因调控网络和预测细胞对干预的反应的主要工具。不幸的是，目前的网络估计算法不能充分地从表达数据中重建基因网络。这并不令人惊讶，因为大多数网络估计算法都是模块化运行的，并且忽略了前面步骤中的不确定性。这项拟议研究的总体目标是：(1)通过使用明确建模并在过程的每一步中纳入不确定性的方法，改进从扰动实验中对基因调控网络的估计，以及(2)使用这些估计的网络来预测细胞对干预的反应。我的长期目标是利用统计学、系统生物学和遗传学领域，对复杂的蜂窝网络进行独立研究。该奖项将提供支持，以获得所需的专业知识，以实现拟议的研究目标并过渡到独立的研究生涯。这将通过课程作业、指导和研究经验的结合来实现。尤其重要的是继续我在分子生物学和癌症基因组学方面的教育，通过正规的课程和基因组实验室技术的指导。这将提供必要的背景，与生物医学研究人员密切合作，开发应对基因组研究前沿挑战的统计方法。定期与我的导师和合作者互动{统计学、计算生物学、生物医学遗传学和癌症研究方面的专家{将为我提供一个丰富的环境，使我能够获得成功过渡到独立研究所需的技能。

DESCRIPTION (provided by applicant): Advances in genomic technology have led to the discovery of numerous genes whose expression differs between cellular conditions; however, genes do not act in isolation, rather they act together in complex networks that drive cellular function. By considering the interactions between genes (and gene products), one gains a more in-depth understanding of the underlying cellular mechanisms. Estimation of these gene regulatory networks is necessary to understand cellular mechanisms, detect differences between cell types, and predict cellular response to interventions. Cancer progression has been shown to produce drastic changes in genetic networks critical to normal cellular function. Some oncogenic mutations produce self-sustaining alterations in the network structure such that removal of the original mutation does not restore normal cellular function. This suggests that identifying the original oncogenic mutation may not be sufficient for a targeted intervention; rather, a detailed understanding of the gene regulatory networks present in both normal and malignant cells may be necessary. Gene perturbation experiments are the primary tool to investigate gene regulatory networks and predict cellular response to interventions. Unfortunately, current network estimation algorithms are unable to adequately reconstruct gene networks from expression data. This is not surprising given that most network estimation algorithms function modularly and disregard uncertainty in previous steps. The overall goals of the proposed research are: (1) to improve the estimation of gene regulatory networks from perturbation experiments, by using methods that explicitly model and incorporate uncertainty in each step of the process, and (2) to use these estimated networks to predict cellular response to intervention. My long term goal is to pursue independent research into complex cellular networks drawing on the fields of statistics, systems biology, and genetics. This Award will provide support to obtain the expertise required to address the proposed research aims and transition to an independent research career. This will be accomplished through a combination of coursework, mentorship, and research experience. Of particular importance is continuing my education in molecular biology and cancer genomics through formal coursework and instruction in genomic laboratory techniques. This will provide the background necessary to work closely with biomedical investigators developing statistical methodology that addresses cutting-edge challenges in genomic research. Regular interaction with my mentors and collaborators {experts in Statistics, Computational Biology, Biomedical Genetics, and Cancer Research {will provide a rich environment in which I can obtain the necessary skills to successfully transition to independent research.