Developing novel algorithms for spatial molecular profiling technologies

开发空间分子分析技术的新算法

• 批准号: 10457848
• 负责人: Guanghua Xiao
• 金额: $ 35.65万
• 依托单位: UT SOUTHWESTERN MEDICAL CENTER
• 依托单位国家: 美国
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-08-01 至 2025-04-30
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10457848
• 关键词: Algorithmic Software; Algorithms; Attention; Bayesian Modeling; Biological; Biological Process; Cell physiology; Cells; Characteristics; Clinical; Communities; Complex; Computational algorithm; Computing Methodologies; Data; Data Analyses; Data Set; Development; Disease; Embryonic Development; Feasibility Studies; Gene Expression; Genes; Genomics; Goals; Graph; Heterogeneity; Image; Image Analysis; Infrastructure; Intuition; Lead; Location; Machine Learning; Messenger RNA; Methodology; Methods; Modeling; Molecular; Molecular Profiling; Morphology; Network-based; Pathologic; Pathway interactions; Patient-Focused Outcomes; Pattern; Phenotype; Physiological; Proteins; Research; Research Personnel; Spatial Distribution; Statistical Methods; Statistical Models; Structure; Technology; Tissue imaging; Tissues; Variant; Visualization; base; bioinformatics tool; biological research; cell type; complex data; computerized tools; data structure; data visualization; deep learning; deep learning algorithm; disease diagnosis; experience; feature selection; flexibility; graph neural network; improved; informatics tool; insight; mental function; novel; preservation; software development; tool; tumor heterogeneity; user friendly software

Project Summary The location, timing and abundance of mRNA and proteins within a tissue underlie the basic molecular mechanisms of cell functions and physiological and pathological developments. For example, the study of expression of thousands of genes simultaneously at different locations could reveal great insights into embryo development, the cooperation of molecular and cellular processes for high-order mental functions, and the molecular basis and clinical impact of intra-tumor heterogeneity. Recent technology breakthroughs in spatial molecular profiling (SMP), including both imaging-based technologies and sequencing-based technologies, have enabled the comprehensive molecular characterization of single cells while preserving their spatial and morphological contexts. Due to the huge potential to deepen our understanding of the molecular mechanisms of cellular and physiological phenotypes, SMP technologies are rapidly gaining attention and a large amount of such data will be generated. However, there are only few computational methods developed to analyze such rich but complex data, and the limitations of computational methods lead to such valuable data being largely under-used. The overarching goal of this study is to develop computational methods to analyze SMP data to characterize detailed molecular spatial distributions and associate such information with cellular phenotypes and physiological phenotypes. The specific aims are as follows: 1. develop novel spatio-statistical methods to characterize spatial distributions of gene expression; 2. develop computational methods to characterize cellular spatial organizations and investigate their relationship with molecular spatial distributions and disease status; 3. develop user-friendly software to facilitate researchers in SMP data analysis and visualization. In order to achieve this goal, we have assembled a strong team with complementary expertise in single-cell genomics, tissue image analysis, spatial modelling, machine learning and software development. If implemented successfully, this platform will greatly facilitate users in understanding molecular and cellular spatial organization in biological tissues and provide comprehensive insights into the underlying biological processes.

项目总结