Data Analysis Core

数据分析核心

基本信息

批准号：
10701828
负责人：
Jeffrey M Spraggins
金额：
$ 57.51万
依托单位：
VANDERBILT UNIVERSITY
依托单位国家：
美国
项目类别：
财政年份：
2022
资助国家：
美国
起止时间：
2022-09-15 至 2026-06-30
项目状态：
未结题

来源：
https://reporter.nih.gov/project-details/10701828
关键词：
3-Dimensional Age Anatomy Atlases Biological Biological Assay Cell Cycle Cells Chemicals Classification Clinical Communities Computer software Cues Data Data Analyses Data Set Data Sources Development Dimensions Dropout Eye Gender Gene Expression Goals Human Human BioMolecular Atlas Program Image Imaging technology Individual Kidney Label Learning Machine Learning Magnetic Resonance Imaging Maps Masks Measurement Medical Imaging Memory Methods Microscopy Mining Mission Modality Modeling Molecular Molecular Analysis Multimodal Imaging Nature Nephrons Organ Pancreas Population Procedures Process Production Proteomics Published Comment Quality Control Reporting Research Resolution Resources Sampling Scheme Spectrometry, Mass, Matrix-Assisted Laser Desorption-Ionization Speed Technology Tissue Donors Tissue Sample Tissues Universities Variant Visualization Work X-Ray Computed Tomography automated segmentation cell type cohort computerized data processing computing resources data integration data mining demographics human tissue imaging facilities insight interest member multimodal data multimodality multiple omics open source spatial integration spatiotemporal tissue mapping transcriptomics

项目摘要

PROJECT SUMMARY – Data Analysis Core. The characterization pipelines built by the VU Biomolecular Multimodal Imaging Center (BIOMIC) Data Analysis Core (DAC) will produce a multimodal molecular atlas of kidney across multiple scales in 2-D and 3-D, comprising rich and varied molecular data from MALDI IMS, MxIF, CODEX, spatial and non-spatial proteomics/ transcriptomics, united by common spatial coordinate space. The DAC will build on our previous developments, including spatial segmentation for both acquisition and analysis, comprehensive molecular identification across modalities, automated data mining across modalities by spatial masks using clinical and temporal cues, and situating data within an “average” human kidney developed from 1000s of medical images. In Aim 1 we will further develop and scale our modality-specific processing methods to prepare the different measurement types for subsequent spatial integration, multimodal analysis, and content/cross-omic relationship mining. Aim 2 will focus on expanding and scaling our characterization pipeline, bringing together the broad array of distinct imaging and -omics measurement types individually processed by the methods in Aim 1. By integrating these varied datasets spatially, temporally, as well as content-wise, we will empirically mine them for cross-modal and molecular-functional relationships. These technologies will establish both 2-D and 3-D multimodal tissue maps that concurrently report hundreds to thousands of biomolecules at cellular resolutions, together with corresponding functional and cell type annotations. In Aim 3 we will compose all spatial and molecular information into a reference atlas by expanding the spatio-temporal mapping of molecular variation, organization, and function beyond single tissue block analyses (as in Aim 2) to multi-block and multi-donor tissue cohorts. Finally, Aim 4 will enable continued coordination with the HuBMAP consortium and HIVE team members by integrating our analyses into the HIVE’s ASCT+B tables for interactive exploration, visualization, and searchable analyses.

项目摘要 - 数据分析核心。 VU 生物分子多模态成像中心 (BIOMIC) 数据分析构建的表征管道 Core (DAC) 将生成跨 2D 和 3D 多个尺度的肾脏多模式分子图谱，包括来自 MALDI IMS、MxIF、CODEX、空间和非空间蛋白质组学的丰富多样的分子数据/ 转录组学，由共同的空间坐标空间统一起来，DAC 将建立在我们之前的开发基础上，包括采集和分析的空间分割、跨区域的全面分子识别模态，使用临床和时间线索通过空间掩模跨模态进行自动数据挖掘，以及在目标 1 中，我们将根据数千张医学图像将数据置于“平均”人类肾脏中。进一步开发和扩展我们特定于模态的处理方法，以准备不同的测量类型用于后续的空间整合、多模态分析和内容/跨组学关系挖掘目标 2。专注于扩展和扩展我们的表征管道，汇集广泛的不同特征通过目标 1 中的方法单独处理成像和组学测量类型。通过集成这些空间、时间和内容方面的各种数据集，我们将凭经验挖掘它们以进行跨模式和这些技术将建立 2D 和 3D 多模式组织图。同时以细胞分辨率报告数百至数千个生物分子，以及在目标 3 中，我们将编写所有空间和分子注释。通过扩展分子变异、组织、其功能超越单一组织块分析（如目标 2），可用于多块和多供体组织队列。最后，目标 4 将通过以下方式实现与 HuBMAP 联盟和 HIVE 团队成员的持续协调：将我们的分析集成到 HIVE 的 ASCT+B 表中，以实现交互式探索、可视化和可搜索分析。