Polly curates and harmonizes datasets from public repositories- processes measurements, links to ontology-backed metadata and transforms them into a unified data model to deliver pristine quality data for faster insights.
Public Repositories Supported
Polly curates datasets of your choice from public repositories, harmonizes them, and delivers in a pristine-quality, ML-ready format, fit for downstream analysis.
Polly retrieves datasets from public repositories and curates and harmonizes data of your choice. Polly's powerful harmonization engine processes measurements, links to ontology-backed metadata and transforms them into a Unified Data Model.
Datasets are mapped with 6 standard fields at dataset and 15 at sample level (can be customized for up to ~30 fields). Commonly used ontologies are MeSH, BRENDA Tissue Ontology, NCBI Taxonomy, Cellosaurus, Cell Ontology and PubChem, for disease, tissue, organism, cell line, cell type, and drug respectively.
Biomedical data harmonized every month.
Data types supported across multi-omics, assay and clinical.
Data sources supported including GEO, PRIDE, CPTAC and more.
Diseases across oncology, metabolic, immunology and more.