domains / families / data / external-model-validation

external-model-validation

separate

Use when validating an existing prognostic risk signature on an external bulk expression cohort with survival outcomes, producing risk scores, Kaplan-Meier curves, risk distribution plots, heatmap, and time-dependent ROC curves. NOT for: model training, feature selection, nomogram construction, calibration analysis, or single-cell data.

When to use

• You already have a fixed prognostic gene signature and coefficients.
• You need to test that model on an independent cohort with bulk expression and survival data.
• You want standard outputs for external validation: risk table, Kaplan-Meier curve, risk score plot, survival status plot, expression heatmap, and time-dependent ROC.

When not to use

• Do not use this skill to train or re-fit a prognostic model.
• Do not use it for nomogram construction, calibration curves, DCA, or diagnostic classification.
• Do not use it for single-cell expression matrices or cohorts without survival endpoints.
• Do not use identifiable patient data without de-identification and local compliance approval.
• Do not use for cohorts with very few events (fewer than 5 events may produce unreliable Kaplan-Meier and ROC results).

Skill page

Source