Informatics Algorithms for Genomic Analysis of Brain Imaging Data

用于脑成像数据基因组分析的信息学算法

• 批准号: 10591596
• 负责人: Jason H. Moore
• 金额: $ 33.53万
• 依托单位: UNIVERSITY OF PENNSYLVANIA
• 依托单位国家: 美国
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-07-01 至 2025-03-31
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10591596
• 关键词: Address; Algorithms; Alzheimer' s Disease; Atlases; Awareness; Beds; Biological; Brain; Brain Diseases; Brain imaging; Characteristics; Collection; Complex; Data; Data Set; Development; Diagnostic; Evaluation; Gene Expression; Genes; Genetic; Genetic Variation; Genetic study; Genome; Genomics; Genotype; Genotype-Tissue Expression Project; Goals; Human; Image; Informatics; Knowledge; Link; Linkage Disequilibrium; Measures; Mental disorders; Methods; Mining; Modeling; Molecular; Multiomic Data; Nature; Neurobiology; Neurologic; Outcome; Pattern; Phenotype; Proteome; Public Health; Quantitative Trait Loci; Research; Software Tools; Source; Structure; System; Systems Biology; Testing; Therapeutic; Time; Tissues; Validation; cohort; data repository; design; epigenome; gene expression database; genetic association; genome-wide; image guided; imaging genetics; in vivo; innovation; insight; knowledge repository; metabolome; multimodality; nervous system disorder; neurobiological mechanism; novel; open source; preference; quantitative imaging; simulation; tool; trait; transcriptome; transcriptomics

Project Summary Brain imaging genetics studies the relationship between genetic variations and brain imaging quantitative traits (QTs) and offers enormous potential to reveal the genetic underpinning of the neurobiological system that can impact the development of diagnostic, therapeutic and preventative approaches for complex brain disorders. Two critical gaps limiting the progress of brain imaging genetics include (1) the unprecedented scale and complexity of the imaging genetic data sets, and (2) lack of intermediate-level omics data to capture the molecular effects linking genetics to brain QTs. Our prior studies have contributed substantially to addressing the first gap. The proposed project will develop new informatics strategies to bridge the second gap, where valuable existing data in the omics domain will be leveraged to link brain imaging and genetics. In this project, we will focus on transcriptomics, and will make use of major transcriptomics data repositories including Genotype-Tissue Expression (GTEx) Project, UK Brain Expression Consortium (UKBEC), and Allen Human Brain Atlas (AHBA). Our overarching goal is to identify brain imaging genetic associations with evidence manifested in the human brain transcriptome. Our hypothesis is that, with additional source of evidence at the transcriptomic level, the identified brain imaging genetic associations are biologically more meaningful and less likely to be false positives. To achieve our goal, we propose four aims. Aim 1 is to develop novel bi-multivariate models incorporating regional tissue-specific expression quantitative trait locus (eQTL) knowledge for mining brain imaging genetic associations. Given that eQTL is a source of tissue-specific evidence to link genotype, gene expression, and brain QTs, we will develop novel eQTL-guided bi-multivariate models to identify imaging genetic associations potentially evidenced by regional tissue-specific eQTL knowledge. Aim 2 is to develop novel bi-multivariate models incorporating brain-wide genome-wide (BWGW) cross-domain co-expression patterns for mining brain imaging genetics associations. AHBA, a BWGW gene expression database, is a natural connection between genome and brain. We propose to develop novel biclustering and bi-multivariate methods to identify meaningful AHBA modules with cross-domain co-expression patterns, and use these patterns to guide the search for co-expression-aware associations between genetic variations and multimodal brain imaging measures. Aim 3 is to develop open source software tools for structure-aware mining of brain imaging genetic associations. Aim 4 is to perform evaluation and validation on both simulated data and real imaging genetics cohorts. Successful completion of the above aims will produce innovative informatics methods and tools for integrative analysis of imaging, genetics and transcriptomics data to address a critical barrier in brain imaging genetics. Using ADNI and related cohorts as test beds, these methods and tools will be shown to have considerable potential for understanding the molecular mechanism of Alzheimer’s disease, and be expected to impact neurological and psychiatric research in general and benefit public health outcomes.

项目概要 脑成像遗传学研究遗传变异与脑成像定量之间的关系 性状（QT）并为揭示神经生物学系统的遗传基础提供了巨大的潜力 可以影响复杂大脑的诊断、治疗和预防方法的发展 限制脑成像遗传学进展的两个关键差距包括（1）前所未有的规模。 和成像遗传数据集的复杂性，以及（2）缺乏中级组学数据来捕获 我们之前的研究对解决遗传学与大脑 QT 之间的分子效应做出了重大贡献。 拟议的项目将制定新的信息学战略来弥补第二个差距。 在该项目中，将利用组学领域的宝贵现有数据将大脑成像和遗传学联系起来。 我们将专注于转录组学，并将利用主要的转录组学数据存储库，包括 基因型组织表达 (GTEx) 项目、英国脑表达联盟 (UKBEC) 和 Allen Human 大脑图谱 (AHBA)。我们的首要目标是通过证据确定大脑成像遗传关联。 我们的假设是，有额外的证据来源。 在转录组水平上，所确定的脑成像遗传关联在生物学上更有意义且更少 为了实现我们的目标，我们提出了四个目标：开发新颖的双多元变量。 结合区域组织特异性表达数量性状基因座 (eQTL) 知识进行挖掘的模型 鉴于 eQTL 是连接基因型的组织特异性证据的来源， 基因表达和大脑 QT，我们将开发新型 eQTL 引导的双多元模型来识别成像 目标 2 是开发区域组织特异性 eQTL 知识可能证明的遗传关联。 结合全脑全基因组（BWGW）跨域共表达的新型双多元模型 AHBA 是一个 BWGW 基因表达数据库，用于挖掘大脑成像遗传学关联的模式。 我们建议开发新的双聚类和双多元变量。 识别具有跨域共表达模式的有意义的 AHBA 模块的方法，并使用这些方法 指导搜索遗传变异和多模式之间的共表达感知关联的模式 目标 3 是开发用于大脑结构感知挖掘的开源软件工具。 目标 4 是对模拟数据和真实数据进行评估和验证。 成功完成上述目标将产生创新的信息学。 对成像、遗传学和转录组学数据进行综合分析的方法和工具，以解决关键问题 使用 ADNI 和相关队列作为试验台，这些方法和工具将克服脑成像遗传学的障碍。 被证明在理解阿尔茨海默病的分子机制方面具有巨大的潜力，并且 预计将影响一般的神经学和精神病学研究，并有益于公共卫生结果。

项目成果

Sparse Interpretation of Graph Convolutional Networks for Multi-Modal Diagnosis of Alzheimer's Disease.

用于阿尔茨海默病多模态诊断的图卷积网络的稀疏解释。

• 发表时间: 2022-09
• 作者: Zhou, Houliang;Zhang, Yu;Chen, Brian Y;Shen, Li;He, Lifang
• 通讯作者: He, Lifang

Multi-Group Tensor Canonical Correlation Analysis.

多组张量典型相关分析。

• 发表时间: 2023-09
• 作者: Zhou, Zhuoping;Tong, Boning;Tarzanagh, Davoud Ataee;Hou, Bojian;Saykin, Andrew J;Long, Qi;Shen, Li
• 通讯作者: Shen, Li

Tensor-Based Multi-Modality Feature Selection and Regression for Alzheimer's Disease Diagnosis.

用于阿尔茨海默病诊断的基于张量的多模态特征选择和回归。

• 发表时间: 2022-10
• 作者: Yu, Jun;Kong, Zhaoming;Zhan, Liang;Shen, Li;He, Lifang
• 通讯作者: He, Lifang

Mediation Analysis and Mixed-Effects Models for the Identification of Stage-specific Imaging Genetics Patterns in Alzheimer's Disease.

用于识别阿尔茨海默病阶段特异性影像遗传学模式的中介分析和混合效应模型。

• 发表时间: 2022-12
• 作者: Pala, Daniele;Lee, Brian;Ning, Xia;Kim, Dokyoon;Shen, Li;ADNI
• 通讯作者: ADNI

Exploring genetic influences on adverse outcome pathways using heuristic simulation and graph data science.

使用启发式模拟和图形数据科学探索遗传对不良结果途径的影响。

• DOI: 10.1016/j.comtox.2023.100261
• 发表时间: 2023-01-01
• 期刊: Computational toxicology
• 作者: Joseph D. Romano;Liang Mei;Jonathan Senn;J. H. Moore;Holly M. Mortensen
• 通讯作者: Holly M. Mortensen