Back

SMART: A Somatic Mutation Annotation and Reporting Tool for cancer genomics

Dominguez, M.; Reddin, I. G.; Gibson, J.; Rudraraju, M.; Veal, K.; Kipps, C.; Williams, A.; Ennis, S.

2026-07-19 bioinformatics
10.64898/2026.07.15.738659 bioRxiv
Show abstract

MotivationTranslational interpretation of somatic variants from targeted oncology panels is hampered by inconsistent transcript prioritisation and by the need for reproducible pipelines that natively integrate OncoKB-derived evidence for research purposes. ResultsWe present SMART (Somatic Mutation Annotation and Reporting Tool), a Dockerised pipeline that embeds OncoKB API-derived annotations, including therapeutic (L1-4), resistance (R1-R3), diagnostic (Dx1-3), prognostic (Px1-3) and FDA levels, directly into a VCF-based workflow. SMART combines this with VEP, CIViC, Cancer Hotspots, ClinVar, SpliceAI, REVEL, LOEUF and gnomAD, applies a unified three-tier transcript prioritisation (whitelist > MANE Select > VEP fallback), and produces three-tiered outputs for computational, bioinformatic and research interpretation. Validation against reference APIs showed full concordance across 804 field-level checks. Availability and ImplementationSource code and Docker image are freely available at https://github.com/WeTGI-colab/SMART under the MIT License. SMART is provided for research use only; use of SMART outputs for patient specific clinical reports, clinical decision-making, or other patient-facing purposes requires appropriate governance and all required third-party licensing, including any OncoKB licence required for patient report generation.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.