Turn cancer screening data into target hypotheses.
Analyzes release-matched cancer screening data to rank gene dependencies and generate biomarker, co-essentiality, and drug-response hypotheses.
When to use it
Use to rank cancer-model dependencies, compare prespecified biomarker cohorts, inspect co-essentiality, or relate genetic dependencies to PRISM compound response.
Give it a cancer-dependency analysis question; it produces analysis outputs accompanied by release metadata, retrieval details, checksums, and source documentation.
Requires Python 3.10 or newer for the bundled analysis helpers.
Requires numpy for the bundled helpers.
Requires pandas for the bundled helpers.
This skill
DepMap
Retrieves DepMap releases and data