6. Interactive analysis with the INCOMMON web app

All the analyses described in the other vignettes (classification, gene mutant dosage, survival analysis and metastatic pattern analysis) are also available without writing any R code, through the INCOMMON web app:

https://ncalonaci.shinyapps.io/incommon/

The app runs the same underlying INCOMMON model used by this package, and is meant for users who want to explore the method interactively or classify a dataset without setting up an R/cmdstanr environment.

With the app you can, briefly:

• upload your own genomic_data and clinical_data tables (in the same format expected by init(), see the Get started vignette) or explore the bundled MSK-MetTropism example cohort;
• run INCOMMON classification of mutation copy number and multiplicity directly from the browser, without installing cmdstanr/Stan locally;
• browse the resulting genome interpretation (gene mutant dosage classes) and the diagnostic/posterior plots produced by the model;
• interactively perform survival and metastatic-tropism analyses on the classified cohort;
• download the classification results and plots for use outside the app.

For full control over the analysis (custom priors, scripting, integration into a larger pipeline), use the R package directly as described in the other vignettes; the app is best suited to quick, exploratory use.