Project-004

Abstract Large-data management systems and computational analysis approaches to immunological research must be able to handle high-throughput, multiparametric assay technologies, which pose unique data management and analysis challenges. Building upon our past work in the Human Immunology Project Consortium (HIPC), including the ImmuneSpace HIPC web portal and data resource, this proposal aims to establish an infrastructure that will streamline data analysis and integration – across research centers and across assay types – across the different scientific projects of this proposal, thus helping achieve more meaningful biological insights into their motivating questions. Our Data Management and Analysis Core (DMAC) will provide data management and associated services required to support the scientific projects of this proposal in three ways. First, we will develop pipelines for importing, annotating, pre-processing, and standardizing immunological assay data to enable data analysis and for deposition into our data management system. These pipelines will utilize existing tools where appropriate; tools will be modified and/or new tools will be developed, when needed. In addition, we will strive to use collaborative, open-source tools. Second, utilizing these pipelines, we will develop a central data management system for data collection, storage, and sharing. The system will be built around the DataPackageR R package that we have developed, and will track data processing as a separate step from statistical analysis. Third, we will provide study design and statistical/computational support for the different research projects of this proposal. These efforts are particularly important for ensuring high sensitivity and specificity to detect e.g. vaccine-induced immune responses and for quantifying these responses with maximal accuracy, reproducibility, and signal-to-noise ratios. By achieving the aims of this grant, unprocessed raw data from diverse labs and immunological assay technologies will be able to be quality controlled, annotated in a standardized fashion, formatted for downstream analysis and modeling, and finally stored on a shared and highly secure infrastructure, providing a seamless and highly reproducible workflow that bolsters the quality of the scientific analyses.