# Cancer AI vocabulary (CanSim terms map)

Source: https://onco.cc/terms/cancer-ai-vocabulary/  
OnCo record `cancer-ai-vocabulary` (Term). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

A hub for the vocabulary of cancer AI: the assays and cohorts models train on, the machine-learning and statistics terms in their papers, the standards their data must follow and the licences that govern reuse.

## Summary

OnCo added this vocabulary on 24 September 2026 from the CanSim terms map, an open list of the terms an open, public-data-first cancer foundation-model programme found itself explaining. The oncology half (assays, data modalities, standards, access rules, clinical concepts) sits in the canonical glossary categories; the methods half sits under Methods and models. Every term paraphrases the page it links, states why a reader meets it in cancer papers, and carries the attribution below. Terms that were only CanSim result names with no general meaning were left out and are listed in docs/CANCERSIM-TERMS-GAP.md. Attribution: CanSim terms map 1.0.0 (docs/onco/terms.json, generated 2026-09-24), CC BY 4.0, attribution: CanSim project, an open, public-data-first cancer foundation-model programme.

## Fields

- Kind: Term
- Last checked: 2026-09-24
- Also known as: cancer AI vocabulary; CanSim terms map; CanSim terms; methods and models glossary
- Tags: cansim-terms; hub

## Notes

- Research vocabulary, not clinical advice: models and metrics described here are research tools unless a page says a product is cleared or approved.

## Connected records

- ideas: [An open foundation model of the cancer cell trained on perturbation data](https://onco.cc/ideas/idea-data-open-cell-foundation-model/), [Patient-level multimodal foundation models for treatment selection](https://onco.cc/ideas/idea-multimodal-foundation-model/)
- collections: [CPTAC (Clinical Proteomic Tumor Analysis Consortium)](https://onco.cc/collections/cptac/), [CZ CELLxGENE / Human Cell Atlas](https://onco.cc/collections/cellxgene-hca/), [DepMap (Cancer Dependency Map)](https://onco.cc/collections/depmap/), [TCGA / NCI Genomic Data Commons](https://onco.cc/collections/tcga-gdc/)
- technologies: [AI compute and model platforms for oncology](https://onco.cc/technologies/ai-compute-platforms/), [Digital pathology & AI](https://onco.cc/technologies/digital-pathology-ai/), [Pathology & radiology foundation models](https://onco.cc/technologies/pathology-foundation-model/), [RNA sequencing & expression profiling](https://onco.cc/technologies/rna-seq/), [Single-cell & spatial profiling](https://onco.cc/technologies/single-cell-spatial/)
- terms: [Ablation study and multi-task heads](https://onco.cc/terms/ablation-study/), [Accuracy, macro-F1 and confusion matrices](https://onco.cc/terms/accuracy-f1/), [Actionable genomic biomarkers](https://onco.cc/terms/cancer-drivers-vs-actionable/), [AJCC stage (TNM staging manual)](https://onco.cc/terms/ajcc-stage/), [Analytical databases as a data catalogue (DuckDB, PostgreSQL)](https://onco.cc/terms/duckdb-catalog/), [Analytical versus clinical validation](https://onco.cc/terms/analytical-vs-clinical-validation/), [AnnData and h5ad files](https://onco.cc/terms/anndata-h5ad/), [Array and table formats: HDF5, Zarr, OME-Zarr, Parquet](https://onco.cc/terms/hdf5-zarr-parquet/), [Attention-based multiple-instance learning (ABMIL, CLAM)](https://onco.cc/terms/abmil/), [Autoregressive (next-token) modelling](https://onco.cc/terms/autoregressive-modelling/), [Batch effects and harmonisation](https://onco.cc/terms/batch-effects/), [Benchmarks, leaderboards and contamination](https://onco.cc/terms/leaderboard-benchmark/), [Bootstrap resampling](https://onco.cc/terms/bootstrap/), [Bulk RNA sequencing (RNA-seq) and the full transcriptome](https://onco.cc/terms/bulk-rna-seq/), [C-index (concordance index), Harrell's and Uno's](https://onco.cc/terms/concordance-index/), [Calibration: reliability diagrams and the Brier score](https://onco.cc/terms/calibration/), [Cell composition confound (tumour versus stroma and immune cells)](https://onco.cc/terms/cell-composition-confound/), [Cell lines as a proxy for patients](https://onco.cc/terms/cell-lines-as-proxy/), [Censoring and events in survival data](https://onco.cc/terms/censoring-and-events/), [CITATION.cff (Citation File Format)](https://onco.cc/terms/citation-cff/), [Clinical covariates (age, stage, nodes, treatment flags)](https://onco.cc/terms/clinical-covariates/), [Clinical text: EHR notes and pathology reports](https://onco.cc/terms/ehr-text-pathology-reports/), [Co-amplification and the 17q12 HER2 amplicon](https://onco.cc/terms/co-amplification/), [Conformal prediction](https://onco.cc/terms/conformal-prediction/), [Contrastive learning (InfoNCE)](https://onco.cc/terms/contrastive-learning/), [Controlled-access genomic data (dbGaP, EGA)](https://onco.cc/terms/controlled-access-data/), [Copy number alteration (CNA)](https://onco.cc/terms/copy-number-variation-term/), [Cross-validation and stratified k-fold](https://onco.cc/terms/cross-validation/), [Data leakage in model evaluation](https://onco.cc/terms/data-leakage/), [Data use agreements and consent for research use](https://onco.cc/terms/data-use-agreements/), [Deep mass-spectrometry proteome (CPTAC)](https://onco.cc/terms/mass-spec-proteome/), [Digital pathology and whole-slide images (WSI)](https://onco.cc/terms/digital-pathology-wsi/), [DNA methylation arrays and beta values (450k, EPIC)](https://onco.cc/terms/methylation-arrays/), [Domain shift and domain adaptation (cell line to patient)](https://onco.cc/terms/domain-adaptation/), [Drug response and drug sensitivity (IC50, AUC)](https://onco.cc/terms/drug-response-sensitivity/), [Drug-response baselines and frameworks: mean-drug floor, LightGBM, DrEval, IMPROVE, DeepTTA](https://onco.cc/terms/drug-response-baselines/), [Drug-response data splits: leave-cell-line-out, leave-drug-out, leave-tissue-out](https://onco.cc/terms/drug-response-splits/), [Embedding (learned representation)](https://onco.cc/terms/embedding/), [Endocrine therapy and endocrine resistance](https://onco.cc/terms/endocrine-therapy-resistance/), [Ensembl gene ID](https://onco.cc/terms/ensembl-gene-id/), [External validation](https://onco.cc/terms/external-validation/), [Fine-tuning versus frozen features, and LoRA](https://onco.cc/terms/fine-tuning-vs-frozen/), [Foundation model](https://onco.cc/terms/foundation-model/), [Gene co-expression structure](https://onco.cc/terms/gene-co-expression/), [Gene set enrichment analysis (GSEA and ssGSEA)](https://onco.cc/terms/gsea/), [Genome builds: GRCh38 versus hg19 (GRCh37)](https://onco.cc/terms/genome-builds/), [Genomic and protein language models: Evo 2, Enformer, ESM](https://onco.cc/terms/genomic-and-protein-language-models/), [GISTIC (copy number driver detection)](https://onco.cc/terms/gistic/), [GPU training and mixed precision](https://onco.cc/terms/mixed-precision-gpu/), [Grade versus stage](https://onco.cc/terms/grade-vs-stage/), [H&E staining (haematoxylin and eosin)](https://onco.cc/terms/h-and-e-staining/), [HGNC gene symbol](https://onco.cc/terms/hgnc-symbol/), [HGVS variant nomenclature](https://onco.cc/terms/hgvs/), [Hugging Face Hub](https://onco.cc/terms/hugging-face-hub/), [Human Phenotype Ontology (HPO)](https://onco.cc/terms/hpo/), [ICD-O-3 (International Classification of Diseases for Oncology)](https://onco.cc/terms/icd-o-3/), [Immunotherapy response and its prediction](https://onco.cc/terms/immunotherapy-response/), [Intra-tumour heterogeneity](https://onco.cc/terms/intra-tumour-heterogeneity/), [Linear probe](https://onco.cc/terms/linear-probe/), [Local-only language models over patient data (privacy by architecture)](https://onco.cc/terms/local-llm-reasoning-layer/), [Logistic regression and nearest-centroid classifiers](https://onco.cc/terms/logistic-regression-term/), [Loss functions: cross-entropy and mean squared error](https://onco.cc/terms/cross-entropy-mse/), [Magnification (20x, 40x) and microns per pixel](https://onco.cc/terms/magnification/), [Masked autoencoders and masked gene modelling](https://onco.cc/terms/masked-modelling/), [Mechanism-of-action recovery and known-biology probes](https://onco.cc/terms/mechanism-of-action-recovery/), [Microarray expression data](https://onco.cc/terms/microarray-expression/), [Model cards and datasheets for datasets](https://onco.cc/terms/model-card/), [Mondo disease ontology](https://onco.cc/terms/mondo/), [mRNA to protein concordance](https://onco.cc/terms/mrna-protein-concordance/), [Multimodal fusion (early, late, modality dropout)](https://onco.cc/terms/multimodal-fusion/), [MutSig (significantly mutated gene detection)](https://onco.cc/terms/mutsig/), [NCI Thesaurus (NCIt)](https://onco.cc/terms/ncit/), [OncoTree cancer classification](https://onco.cc/terms/oncotree-term/), [Open licences: Apache-2.0, MIT and CC BY 4.0](https://onco.cc/terms/open-licences/), [Open weights, open code and gated models](https://onco.cc/terms/open-weights/), [Out-of-distribution detection (Mahalanobis guard)](https://onco.cc/terms/ood-detection/), [Pathology foundation models: UNI, UNI2, Virchow2, CTransPath, CONCH, TITAN](https://onco.cc/terms/pathology-foundation-models/), [Pathway activation state (phosphosignalling)](https://onco.cc/terms/pathway-activation-state/), [Permutation test and the Mann-Whitney U test](https://onco.cc/terms/permutation-test/), [Pharmacogenomics](https://onco.cc/terms/pharmacogenomics-term/), [Post-transcriptional and post-translational regulation](https://onco.cc/terms/post-transcriptional-regulation-term/), [Pre-registered experiment](https://onco.cc/terms/pre-registered-experiment/), [Principal component analysis (PCA) as a feature compressor](https://onco.cc/terms/pca/), [Provenance fields for research data](https://onco.cc/terms/provenance-fields/), [Quantile normalisation, rank transforms and z-scores](https://onco.cc/terms/quantile-normalisation/), [Radiology imaging as a data modality (CT, MRI, TCIA)](https://onco.cc/terms/radiology-imaging-modality/), [Reproducibility and negative results](https://onco.cc/terms/reproducibility/), [Research use only (RUO)](https://onco.cc/terms/research-use-only/), [Reverse-phase protein array (RPPA)](https://onco.cc/terms/rppa/), [Ridge regression and the multilayer perceptron](https://onco.cc/terms/ridge-regression/), [Risk score and risk percentile](https://onco.cc/terms/prognosis-risk-percentile/), [ROC-AUC, PR-AUC and time-dependent AUC](https://onco.cc/terms/roc-auc/), [RxNorm drug nomenclature](https://onco.cc/terms/rxnorm/), [Self-supervised pretraining (SSL)](https://onco.cc/terms/self-supervised-pretraining/), [Single-cell and transcriptome foundation models: UCE, GeneCompass, BulkFormer, BulkRNABert](https://onco.cc/terms/single-cell-foundation-models/), [Single-cell RNA sequencing (scRNA-seq, 10x Chromium)](https://onco.cc/terms/single-cell-rna-seq/), [Software as a medical device (SaMD)](https://onco.cc/terms/samd/), [Somatic mutations from exome and genome sequencing (WXS, WGS)](https://onco.cc/terms/somatic-mutations-wxs-wgs/), [Spatial autocorrelation (Moran's I)](https://onco.cc/terms/spatial-autocorrelation/), [Spatial transcriptomics platforms (Visium HD, Xenium, MERFISH, CosMx, CODEX)](https://onco.cc/terms/spatial-transcriptomics-platforms/), [Spatially aware clustering (SpaGCN, KNN smoothing)](https://onco.cc/terms/spagcn/), [Spearman rank correlation](https://onco.cc/terms/spearman-correlation/), [STAR and Salmon (RNA-seq alignment and quantification)](https://onco.cc/terms/star-salmon/), [Stroma-rich and desmoplastic tumours in molecular data](https://onco.cc/terms/desmoplastic-stroma-rich/), [Survival outcomes as time plus event (OS, PFS)](https://onco.cc/terms/os-pfs-time-event/), [Targeted panel sequencing](https://onco.cc/terms/targeted-panel-sequencing/), [TCGA barcode](https://onco.cc/terms/tcga-barcode/), [TCGA open versus controlled data tiers](https://onco.cc/terms/tcga-tiers/), [Tertiary lymphoid structures (TLS)](https://onco.cc/terms/tertiary-lymphoid-structures/), [Tile and patch encoding of slides](https://onco.cc/terms/tile-patch-encoding/), [Tissue-of-origin signal in tumour data](https://onco.cc/terms/organ-of-origin-signal/), [Tokenisation (genes, tiles and sequence as tokens)](https://onco.cc/terms/tokenisation/), [TPM, FPKM and raw counts (expression units)](https://onco.cc/terms/tpm-fpkm-counts/), [Train, validation and test split discipline](https://onco.cc/terms/train-test-discipline/), [Transfer learning and the low-label regime](https://onco.cc/terms/transfer-learning/), [Transformer and attention](https://onco.cc/terms/transformer-architecture/), [Tumour board (multidisciplinary team meeting)](https://onco.cc/terms/tumour-board/), [Tumour evolution (somatic evolution)](https://onco.cc/terms/tumour-evolution/), [Tumour purity](https://onco.cc/terms/tumour-purity/), [Uberon anatomy ontology and the Cell Ontology](https://onco.cc/terms/uberon/), [Uncertainty quantification and confidence gates](https://onco.cc/terms/uncertainty-quantification/), [Units of measurement ontology (UO)](https://onco.cc/terms/units-ontology/), [Univariate Cox scores and time-dependent metrics](https://onco.cc/terms/time-dependent-auc/), [Variant calling](https://onco.cc/terms/variant-calling/), [Variant effect prediction](https://onco.cc/terms/variant-effect-prediction/), [Virtual cell models and in-silico perturbation screens](https://onco.cc/terms/virtual-cell-models/), [Zenodo DOIs for data and code](https://onco.cc/terms/zenodo-doi/), [Zero-shot prediction](https://onco.cc/terms/zero-shot/)
- targets: [DCLK1](https://onco.cc/targets/dclk1/), [DNAJC12](https://onco.cc/targets/dnajc12/), [E2F7](https://onco.cc/targets/e2f7/), [GRB7](https://onco.cc/targets/grb7/), [MIEN1](https://onco.cc/targets/mien1/), [PGAP3](https://onco.cc/targets/pgap3/), [PHLDA1](https://onco.cc/targets/phlda1/), [PHLPP2](https://onco.cc/targets/phlpp2/), [PNMT](https://onco.cc/targets/pnmt/), [SPRY4](https://onco.cc/targets/spry4/), [STARD3](https://onco.cc/targets/stard3/)

---
JSON: https://onco.cc/api/v1/entities/cancer-ai-vocabulary.json