Skip to content

Precision Oncology

CAPRI

Evidence-ranked variant interpretation, in one workbench.

Cancer Actionability & Precision Response Interpreter. CAPRI unifies twenty-nine biomedical sources into a single workbench — normalized variants, AMP/ASCO/CAP-style Tier I–IV actionability, structure-aware drug context, cohort analytics, and de-identified tumor-board summaries.

Sign-up is open — email verification or Google sign-in. No institutional affiliation required.

510,504

Variants in the live catalogue

1,084

Genes in the live catalogue

29

Integrated biomedical sources

Counts read live from capri.celaron.com/api/stats on 25 October 2026; capability list verified against the public API schema.

CAPRI

Why it matters.

Interpreting a cancer genome means reconciling dozens of knowledgebases, each with its own identifiers, evidence levels, and blind spots. CAPRI normalizes variants on upload and ranks what it finds by curated clinical evidence, so a Tier I finding is never buried under a thousand Tier III variants.

The workbench covers the full interpretation arc: cohort-level analytics, structure and docking context, drug and trial matching, population frequencies including the Qatar Genome Programme, and printable tumor-board reports. It was developed from research at QCRI / HBKU Precision Medicine in Summer 2026 and presented as an A1 research poster.

Capabilities

What it does.

Twenty integrated data sources, one workbench — from raw VCF upload to a printable tumor-board report.
  • Unified search across genes, variants, drugs, and diseases, with transparent evidence ranking.
  • Upload VCF, MAF, or TSV cohorts; variants are validated, normalized, and quality-checked before analysis.
  • Curated clinical evidence from CIViC and CGI with AMP-style Tier I–IV actionability and A–E evidence levels.
  • Structure context: PDB mappings, residue-level annotation, pocket structures, and docking jobs with pose SDF output.
  • Cohort analytics: oncoprint, lollipop, TMB, actionability funnel, mutation co-occurrence, and Benjamini–Hochberg-corrected pathway enrichment.
  • Drug intelligence from ChEMBL, DrugBank, DGIdb, GDSC, and Cell Model Passports, with sensitivity and interaction context.
  • Population context from gnomAD and the Qatar Genome Programme, including local enrichment analysis.
  • Tumor-board-style actionability summaries as printable reports, with batch report generation.

Provenance

Research-grade, and open to sign-up.

Built from research at QCRI / HBKU Precision Medicine (Summer 2026), presented as an A1 research poster.

Sign-up is open — email verification or Google sign-in. No institutional affiliation required.

Create an account and start interpreting.

Sign-up is open — email verification or Google sign-in. No institutional affiliation required.