National Library of Medicine 2R01LM013863-05

Human Microbiome Compendium: large-scale curation and processing of human microbiome datasets

The project expands the Human Microbiome Compendium by integrating over one million microbiome samples into a detailed, searchable dataset using machine learning. It aims to standardize metadata, evaluate microbiome-host relationships, and develop tools including a web app for enhanced microbiome research and data sharing.

Description

This project aims to expand the Human Microbiome Compendium by integrating more than one million samples of human microbiome data into a unified dataset that characterizes sample content and host information at a level far more specific and searchable than that of the NCBI databases in which they're stored. Researchers will be able to retrieve and analyze sequence-level information about the microbial composition of each sample, paired with information built using state-of-the-art machine learning techniques. The objectives of this proposal are consistent with the NOFO for curation at scale for biomedical research assets—in addition to generating the largest dataset of its kind for human microbiome data, this project will also develop automation approaches and software tools that will enable others to build compendia of their own using publicly available data.

View original record