Gradient Biotech

Overview

Oncology workflows connect molecular data to clinical outcomes in a single study workspace. Each use case below maps a common research question to the platform capabilities, data requirements, and suggested analysis path.

Use case index

Use caseResearch questionKey capabilities
Immuno-oncology responseWhy do some patients respond to checkpoint blockade and others do not?Immune deconvolution, TIDE phenotype, exhaustion scoring, cell-cell communication, repertoire analysis
Mutation to outcomesWhich alterations and mutational features associate with survival or response?Mutation landscape, TMB, signatures, oncoprint, survival analysis
Malignant cell CNV evidenceWhich malignant populations and CNV-defined subclones shape tumor biology and downstream TME signals?CompBio CNV handoff, malignant-cell detection, CNV event evidence, mutation overlap, survival/response feature export
TME and signalingWhat cell populations interact and which signaling axes dominate the TME?TME composition, ligand-receptor scoring, pathway aggregation, condition comparison
Spatial TME communicationWhich tumor regions and boundaries concentrate immuno-oncology signaling?Spatial TME regions, boundary priorities, communication maps
Integrated cohort evidenceWhich signals repeatedly support a responder, survival, mutation, or treatment-arm hypothesis across completed analyses?Cohort comparison, multi-run evidence, q-values, source-run traceability
Multi-omic integrationHow do transcriptomic, mutational, and clinical data combine in one cohort?Full study workflow across TME, immune, mutation, survival, and interpret
Predictive biomarker discoveryCan expression features predict a clinical outcome out-of-sample, and which genes carry the signal?Cross-validated random-forest modeling, permutation testing, top-biomarker ranking

Who these use cases serve

AudienceTypical goals
Tumor biology researchersCell-cell communication, spatial TME mapping, immune infiltration, cohort comparison
Translational oncology teamsMutation landscape, therapy-response correlation, survival analysis, biomarker endpoints
Immuno-oncology researchersImmune deconvolution, exhaustion profiling, TCR/BCR repertoire, IO response prediction
Computational oncology groupsReproducible multi-omic pipelines, cohort integration, publication-ready outputs
Pharma and biotech programsMulti-omic patient stratification, drug response correlation, biomarker endpoint analysis
Clinical trial teamsOmics-to-outcomes integration, longitudinal cohort analysis, provenance for review

Common data requirements

Most use cases start with an oncology study containing:

  • Samples with patient/sample keys, timepoints, and group labels (responder, non-responder, treatment arm)
  • Clinical endpoints for survival time, event status, and response
  • Molecular inputs — expression matrices, MAF files, and optional repertoire data

Single-cell and spatial data are ingested through the Computational Biology area and linked to oncology samples. Whole-slide images are ingested through Pathology when tissue morphology context is needed.

Choosing a starting point

If your primary data is…Start with…
Bulk RNA-seq from treated cohortImmuno-oncology response
Tumor sequencing (MAF/VCF)Mutation to outcomes
CompBio single-cell CNV handoffMalignant cell CNV evidence
Single-cell RNA-seqTME and signaling
Spatial coordinates + cell labelsSpatial TME communication
Multiple modalities + clinical outcomesMulti-omic integration
Multiple completed oncology runsIntegrated cohort evidence
Expression + a labeled outcome to predictPredictive biomarker discovery

What these use cases are not

These workflows support exploratory and translational research. They are not clinical diagnostic or treatment decision workflows. All AI interpretation outputs include research disclaimers and cite computed metrics only.

Next steps