BEACON

BEACON identifies expression-driven genetic dependencies by integrating transcriptomic, proteomic, and genetic dependency data to prioritize therapeutic targets in cancer.

Key Features:

• Bayesian Framework: Uses a Bayesian approach to identify correlations between high gene expression and increased vulnerability upon gene knockout, implemented with JAGS and MCMC simulations.
• Integration of Multi-Omics Data: Combines transcriptomic and proteomic profiles with genetic dependency data across 17 tissue lineages to assess expression-dependent vulnerabilities.
• Identification of Druggable Targets: Detects established drug targets such as BCL2, ERBB2, EGFR, ESR1, and MYC and highlights genes with both mRNA- and protein-expression driven dependency as candidate targets.
• Enrichment for Approved Drug Targets: Shows significant enrichment for approved drug targets (3.8-fold) and for druggable oncology targets (7- to 10-fold).
• Experimental Validation: Predictions have been experimentally validated, including genes GRHL2, TP63, and PAX5 whose depletion reduces tumor cell growth and survival in dependent cells.

Scientific Applications:

• Novel target discovery: Identification of new therapeutic targets by revealing expression-driven dependencies, including in tumors lacking druggable mutations.
• Target prioritization for precision oncology: Prioritization of candidate targets for targeted therapies based on integrated expression and dependency evidence.
• Cross-lineage vulnerability analysis: Comparative assessment of gene function and vulnerabilities across 17 tissue lineages to reveal lineage-specific targets.

Methodology:

BEACON integrates transcriptomic and proteomic profiles with genetic dependency data across 17 tissue lineages and applies a Bayesian statistical model implemented in JAGS with MCMC to infer correlations between gene expression and dependency (expression-driven dependency).