Welcome

The Oncology area of Gradient Biotech is a web workspace for multi-omic, spatial, and clinical oncology research. It builds on the Computational Biology foundation — single-cell, spatial, bulk RNA-seq, and biomarker pipelines — and extends it with oncology-specific capabilities: cell-cell communication, immune profiling, mutation landscape analysis, and clinical outcome integration.

What you can do today

Capability	Status
Study and cohort management	Tumor type, stage, treatment arm, response status, sample metadata
Clinical endpoints	Survival, recurrence, and response records linked to patients
Mutation datasets	MAF/VCF ingestion records with per-sample variant tables
Repertoire datasets	TCR/BCR sequencing file records for clonotype analysis
Tumor microenvironment	TME composition view (calls compbio single-cell infrastructure)
Cell-cell communication	Ligand-receptor scoring, pathway aggregation, condition comparison
Immuno-oncology profiling	Deconvolution, TIDE-like phenotype, exhaustion, repertoire metrics
Mutation landscape	TMB, signatures, oncoprint, copy number, driver pathway enrichment
Survival analysis	Kaplan-Meier, log-rank testing, Cox regression, longitudinal trajectories
AI interpretation	Grounded summaries for communication, immune, mutation, and survival outputs
Run history	Study-level pipeline run tracking with artifact links

How the product is organized

Every analysis lives inside a study — your oncology project container. From the oncology dashboard you create or open a study, then work through:

1. Overview — study metadata, workflow cards, and run history
2. Data — samples, clinical endpoints, mutation and repertoire datasets
3. TME — tumor microenvironment composition (compbio-backed)
4. Communication — ligand-receptor and signaling network analysis
5. Immune — deconvolution, immune phenotype, exhaustion, and repertoire profiling
6. Mutations — somatic mutation landscape and oncoprint summaries
7. Survival — clinical endpoint analysis stratified by molecular features
8. Interpret — AI-assisted summaries grounded in completed pipeline outputs

Pipeline jobs run asynchronously. Launch jobs from workflow pages and refresh when complete.

Who this is for

• Tumor biology researchers studying cell-cell communication, spatial TME mapping, and immune infiltration
• Translational oncology teams correlating mutation landscapes with therapy response and survival
• Immuno-oncology researchers profiling exhaustion states, repertoire diversity, and IO response phenotypes
• Computational oncology groups running reproducible multi-omic pipelines with full provenance

What this is not

• Not a regulated clinical diagnostic or clinical decision support tool
• Not a replacement for established algorithms — it orchestrates and modernizes them
• Not a standalone single-modality tool — value compounds as more data types connect within a study

Next steps

• Use cases — scenario guides for IO response, mutation-outcome, TME signaling, and multi-omic integration
• Quick start — create a study and run your first oncology pipeline
• Key concepts — studies, samples, runs, and artifacts
• Study workflow — how the main sections fit together