Computing to Support High-Throughput and High Content Imaging Assays

代表保罗·松代拉教授向怀特黑德生物医学研究所授予了一笔赠款，用于为与成像和系统生物学研究相关的数据存储硬件提供部分支持。在典型的研究中，图像数据由高内容或高分辨率成像平台获取，由共享的多处理器计算机处理，组织在图像数据库中，并存储在硬盘或磁带系统中。然而，自动化成像项目能够每月生成几TB的图像数据，但受到低成本但易于访问的足够容量的档案存储的限制。有了这项NSF奖，WI调查人员将能够积累数TB级的图像数据集，这些数据集将存储在固定内容数据存储系统中，用于下游处理和分析。Wi研究人员研究大规模系统层面的问题，包括细胞迁移、细胞新陈代谢、细胞壁合成、蛋白质折叠和通过膜受体传递信号。对驱动这些过程的蛋白质组机制的系统研究需要收集大容量的GB-TB图像数据集，并对图像中的静态和动态特征进行量化。对这些数据集的分析揭示了系统水平的趋势、以前未知基因的细胞功能以及细胞过程的动力学和机制。收集、组织和存储结核病规模的图像数据集的问题吸引了与不同行业合作伙伴的合作。图像数据库、图像数据集的存储、大规模图像处理和分析、高端成像和基于成像的生物信息学是女性和少数民族学生WI实验室培训以及麻省理工学院和其他大学附属研究生和博士后培训的热门研究课题。

A grant has been awarded to the Whitehead Institute for Biomedical Research on the behalf of Prof. Paul Matsudaira for partial support of data storage hardware associated with imaging and systems biology research. In typical studies, image data is acquired by high content or high resolution imaging platforms, processed by shared multi-processor computers, organized in image databases, and stored on hard disk or tape systems. However, automated imaging projects are capable of generating several Tbs per month of image data but are limited by sufficient, capacity of low cost but easily accessed archival storage. With this NSF award WI investigators will be able to accumulate multi-TB-scale image datasets that will be stored on a fixed content data storage system for downstream processing and analysis. WI investigators study large-scale systems-level problems in cell migration, cell metabolism, cell wall synthesis, protein folding, and signaling through membrane receptors. Systems-based studies of the proteome machinery that drive these processes requires the collection of large GB-TB size image datasets and quantitation of the static and dynamic features in images. Analyses of these datasets reveal systems-level trends, cellular functions for previously unknown genes, and the kinetics and mechanics of cellular processes. The problem of collecting, organizing, and storing TB-scale image datasets attracts collaboration with diverse industrial partners. The image databases, the storage of image datasets, large-scale image processing and analysis, high-end imaging, and imaging-based bioinformatics are attractive research problems for the training in the WI labs of women and minority students and for the training of graduate students and postdocs affiliated with MIT and other universities.