CTA decision and sequence-role contract
Read the illustrated, searchable report · Download the PDF · Current report receipt · Full report data.
Current selection policy: oncoref.cta-selection.2026-11-08.1. Consumer-role policy:
oncoref.sequence-roles.2026-11-04.1. These identities are separate because changing
uncertainty handling need not change a biological classification.
All 19,893 candidate protein records enter classification. Translation support is an admission requirement; 295 records lack the former coding/established-ncORF entry evidence and remain visible as Unsafe or Other. No record is removed before the normal-evidence assessment. The classified results are 607 CTA, 18,259 Unsafe and 1,027 Other. Identical selected reference sequences count together; member gene loci remain traceable. These are representative proteins, not a complete inventory of splice isoforms or modified proteoforms. Inventory and translation admission.
An OncoRef CTA candidate is a reference-resolved protein with supported cancer expression and qualifying reproductive or developmental association that passes the named normal-tissue evidence, coverage and selection rules. CTA admission requires qualifying coding annotation or supported ncORF evidence and a complete, approved MANE Select or Ensembl canonical representative, with any conflicting coding evidence resolved. Every input record is assessed, including those with insufficient translation support.
Evidence status and consumer roles
The categories alone do not determine allowance. An Other:Nonreproductive record
can also have incomplete normal evidence. Check assessment_status, identity and
the explicit sequence role. Unsafe exclusion takes precedence; otherwise unresolved
assessments require review. Only assessed, identity-resolved records receive allow.
Allowance means no exclusion from that record under this policy; other occurrences
can still exclude shared sequence.
| Role | Proteins |
|---|---|
| Exclude | 18,259 |
| Review | 605 |
| Allow under this policy | 1,029 |
Compared with the pinned 1.29.0 snapshot, 653 classifications and 768 roles change. Every changed record retains baseline/current values. The full role table preserves identity, reasons and policy.
Public API
from oncoref import cta_selection
definition = cta_selection.contract()
gates = cta_selection.gates()
roles = cta_selection.sequence_dispositions()
coverage = cta_selection.coverage()
The gate catalog identifies independent questions; it does not claim every assay is measured. The current decision audit retains source observations and overlapping reasons. Definitions, coverage and reproduction receipt are machine readable.
These are representative-protein decisions. Translation qualification and the verified-record loader retain usable translation records, original sequences and all source occurrences. A gene mapping, CTA label or exact sequence match does not validate every isoform or an approximate patient-HLA similarity threshold. Existing historical FASTA bundles keep their own policy and reference identities.