Uniform processing and analysis of IGVF massively parallel reporter assay data with MPRAsnakeflow

Area of Science:

• Genomics
• Regulatory Genomics
• Functional Genomics

Background:

• Interpreting human genomic variations, especially non-coding ones, is challenging due to limited functional annotation.
• Experimental validation is essential for understanding the functional impact of genetic alterations on phenotypes.
• Massively Parallel Reporter Assays (MPRAs) offer high-throughput quantification of regulatory element activity and variant effects.

Purpose of the Study:

• To address challenges in MPRA data integration, reproducibility, and meta-analysis caused by diverse designs and processing parameters.
• To establish community standards, including harmonized file formats and analysis pipelines for MPRAs.
• To present computational tools (MPRAlib and MPRAsnakeflow) for uniform MPRA data processing.

Main Methods:

• Development of community standards and harmonized file formats for MPRA data.
• Creation of computational tools (MPRAlib and MPRAsnakeflow) for processing MPRA data from raw reads to visualization.
• Characterization of technical variability sources in MPRA data using diverse datasets.

Main Results:

• Established standardized formats and robust analysis pipelines for various MPRA library types and experimental designs.
• Demonstrated uniform processing of MPRA data, enabling better integration and reproducibility.
• Identified and characterized technical variability, including barcode sequence bias and delivery method effects.

Conclusions:

• The developed standards and tools facilitate robust and reproducible MPRA research.
• Best practices for MPRA data generation and analysis are established.
• Publicly available tools and standards provide a foundation for collaborative regulatory genomics research.