What problem does it solve?
Interpreting CRISPR-Cas9 genetic screens requires chaining together sgRNA count processing, quality control, gene-level essentiality scoring, pathway enrichment, and druggability assessment — a workflow that is error-prone and slow when done manually. This Skill provides a complete 8-phase analysis pipeline that turns raw sgRNA count matrices or gene hit lists into prioritized, clinically contextualized therapeutic target reports.
Core Features & Use Cases
- End-to-end screen analysis: Load sgRNA count matrices, run QC (library size, Gini coefficient, low-count filtering), normalize counts, compute log2 fold changes, and score genes with MAGeCK-like RRA or BAGEL-like Bayes Factor methods using bundled CEGv2/NEGv1 reference gene sets.
- Synthetic lethality and target discovery: Compare essentiality between wildtype and mutant contexts, then prioritize hits by integrating essentiality, expression, and DGIdb druggability into a composite priority score.
- ToolUniverse integration: Enrich hits via Enrichr/Reactome, build STRING PPI networks, query PubMed, Pharos, and clinical trials, with a documented Pharos/Open Targets fallback when DepMap APIs are unavailable.
- Use Case: Given 20 hits from an A549 lung cancer dropout screen, the skill classifies essentiality, finds cell-cycle checkpoint enrichment, and ranks KRAS, EGFR, and WEE1 as top targets with validation recommendations.
Quick Start
Ask the AI to analyze your CRISPR screen gene list, for example: "Analyze these CRISPR dropout screen hits from A549 lung cancer cells and generate a target prioritization report."
