FILER: a framework for harmonizing and querying large-scale functional genomics knowledge

Area of Science:

• Genomics
• Bioinformatics
• Data Science

Background:

• High-throughput genomic analysis requires querying and integrating heterogeneous data from various sources.
• Data complexity arises from diverse experimental assays, biological contexts, and file formats, hindering efficient analysis.

Purpose of the Study:

• To introduce FILER (FunctIonaL gEnomics Repository), a framework designed to overcome challenges in querying large-scale functional genomics and annotation data.
• To provide a unified, scalable interface for accessing and analyzing harmonized genomic datasets.

Main Methods:

• Developed FILER, a framework integrating over 50,000 harmonized genomic datasets from more than 20 sources.
• Implemented a scalable genomic search and querying interface supporting GRCh37/hg19 and GRCh38/hg38.
• Benchmarked FILER's scalability by querying 7 x 10^9 hg19 records, demonstrating sub-linear query time increase with a 1000-fold query volume increase.

Main Results:

• FILER offers streamlined access to >50,000 harmonized datasets across >20 sources, >1100 tissues/cell types, and >20 experimental assays.
• The framework supports querying >17 billion genomic records for both GRCh37/hg19 and GRCh38/hg38.
• Performance benchmarks confirm FILER's high scalability, with query times increasing sub-linearly as query volume grows.

Conclusions:

• FILER facilitates reproducible research by simplifying the integration and querying of large-scale genomic data.
• The framework's scalability and comprehensive data integration streamline complex genomic and genetic analytical workflows.
• FILER is available for cloud or local deployment, enabling integration with custom bioinformatics pipelines.