Using individual barcodes to increase quantification power of massively parallel reporter assays

Area of Science:

• Genomics
• Bioinformatics
• Molecular Biology

Background:

• Massively parallel reporter assays (MPRAs) measure regulatory sequence activity using pooled sequence-tagged reporter genes.
• Existing MPRA analysis tools like MPRAnalyze and mpralm have limitations in speed, statistical power, and robustness to outliers.
• MPRA data analysis relies on modeling count data from reporter gene constructs.

Purpose of the Study:

• To develop a novel MPRA analysis framework, BCalm, to overcome limitations of current tools.
• To enhance statistical power and robustness in MPRA data analysis.
• To provide a fast, accurate, and user-friendly tool for MPRA data interpretation.

Main Methods:

• BCalm adapts the mpralm framework by modeling individual barcode counts instead of aggregated counts per sequence.
• The method was evaluated on simulated MPRA data and a large-scale lentiviral MPRA library.
• BCalm incorporates functionalities for input file preparation and analysis of sequence-specific enhancing or repressing activities.

Main Results:

• BCalm demonstrates improved statistical power and robustness to outliers compared to existing methods.
• The framework shows superior performance on both simulated and real MPRA datasets.
• BCalm provides built-in plotting functionalities for straightforward results interpretation.

Conclusions:

• BCalm offers a robust and accurate tool for Massively Parallel Reporter Assay data analysis.
• The new framework addresses key limitations of previous MPRA analysis tools.
• BCalm is available as an open-source package for the research community.