BAGEL2

BAGEL2 identifies gene essentiality from CRISPR-Cas9 genome-wide pooled loss-of-function screens using a Bayesian framework to enable robust detection of essential genes and cancer-relevant targets.

Key Features:

• Improved Model with Greater Dynamic Range: Employs an updated Bayesian model that yields a greater dynamic range of Bayes Factors, increasing sensitivity for detecting tumor suppressor genes.
• Multi-Target Correction: Incorporates a multi-target correction to reduce false positives caused by off-target effects of CRISPR guide RNAs.
• Cross-Validation Strategy: Implements a cross-validation strategy that improves performance approximately tenfold compared with previous bootstrap resampling approaches.
• Screen Quality Metric: Computes a replicate-level screen quality metric to assess data reliability and to characterize algorithm behavior on lower-quality data.

Scientific Applications:

• Functional Genomics: Enables precise identification of essential genes from CRISPR-Cas9 pooled loss-of-function screens.
• Cancer Target Discovery: Facilitates detection of tumor suppressor genes and prioritization of cancer-relevant dependency candidates.
• Genome-wide Loss-of-Function Studies: Applicable to analysis of genome-wide CRISPR knockout fitness screens for mapping gene essentiality.
• Off-target Robustness Assessment: Supports assessment and mitigation of off-target effects from guide RNAs during essentiality classification.

Methodology:

BAGEL2 uses a Bayesian framework that computes Bayes Factors for gene essentiality, incorporates multi-target correction, applies cross-validation (replacing bootstrap resampling), and computes a replicate-level screen quality metric.