ScreenBEAM: a novel meta-analysis algorithm for functional genomics screens via Bayesian hierarchical modeling.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Functional genomics (FG) screens, utilizing RNAi or CRISPR, are crucial for genome-wide loss-of-function studies in therapeutic target discovery.
• High-throughput FG screening data analysis faces challenges due to off-target effects, reagent variability, and experimental noise, complicating the integration of multiple guides per gene.
• Rigorous statistical analysis is essential to mitigate false positives in large-scale FG screening data.

Purpose of the Study:

• To develop and evaluate a novel meta-analysis approach for functional genomics screens using Bayesian hierarchical modeling.
• To introduce the Screening Bayesian Evaluation and Analysis Method (ScreenBEAM) for robust analysis of FG screening data.

Main Methods:

• Utilized publicly available large RNAi and CRISPR repositories.
• Applied a novel meta-analysis approach via Bayesian hierarchical modeling, termed ScreenBEAM.
• Evaluated ScreenBEAM's performance against classical algorithms and next-generation sequencing-focused approaches.

Main Results:

• ScreenBEAM robustly outperforms classical microarray algorithms and recent next-generation sequencing approaches.
• The proposed meta-analysis strategy effectively integrates all available data for improved FG screen analysis.
• ScreenBEAM demonstrates strong performance even with low-quality functional genomics screening data, common in public datasets.

Conclusions:

• ScreenBEAM offers a robust and effective meta-analysis strategy for functional genomics screens.
• The method is particularly valuable for analyzing the vast majority of public FG datasets, which often exhibit lower quality.
• This approach enhances the reliability of therapeutic target discovery from large-scale functional genomics studies.