Gradient Biotech
U

Menu

Study workflow

Oncology follows a left-to-right research flow connecting molecular data to clinical outcomes. Sample and clinical metadata should be registered before launching analysis pipelines.

Create study → Register samples & endpoints → Upload mutation/repertoire data → Run oncology pipelines → Interpret → Report

Dashboard — area home

The oncology dashboard (/areas/oncology/dashboard) shows recent studies and links to create a new study.

Readiness signal: at least one study with registered samples and one completed analysis run.

Study overview — project hub

The study page (/areas/oncology/studies/{id}) is your project hub:

  • Study metadata — tumor type, stage, treatment arm, response status
  • Workflow cards — links to TME, communication, immune, mutations, survival, and interpret pages
  • Run history — all pipeline runs for this study

Data — cohort organization

The data page (/areas/oncology/studies/{id}/data) manages:

  • Sample records with patient/sample keys and timepoints
  • Clinical endpoint records for survival and response
  • Mutation and repertoire dataset registrations

See Data upload.

Analysis workflow pages

Each analysis type has a dedicated page under the study:

PageRoutePurpose
TME/tmeTumor microenvironment composition (compbio-backed)
Communication/communicationLigand-receptor and signaling network analysis
Immune/immuneImmuno-oncology profiling and repertoire metrics
Mutations/mutationsSomatic mutation landscape and oncoprint
Survival/survivalKaplan-Meier, log-rank, and Cox regression
Interpret/interpretAI-assisted summaries of completed runs

See the Analysis guides section for pipeline details on each page.

Typical paths

Study typePath
Mutation-first cohortRegister samples → upload MAF → mutation landscape → survival stratified by TMB
IO responder analysisRegister samples with response labels → immune profile → cell communication with condition comparison → interpret
Multi-omic integrationTME (compbio) → communication → immune profile → mutation landscape → survival → interpret
Translational endpointClinical endpoints → survival with molecular feature stratification → interpret

Integration with other areas

  • Computational Biology — single-cell TME profiling, spatial transcriptomics, differential expression, biomarker ML
  • Pathology — whole-slide imaging for pathology-integrated spatial TME context

Oncology consumes these capabilities rather than duplicating them. See products/oncology/product.md for the full integration map.

Next steps