← Mutation landscapeJSON

Neuroblastoma mutation landscape report

BioTransfer local release · 2026-09-09 · Source status: PUBLIC SOURCE SNAPSHOT

Executive summary

This report documents the first local Neuroblastoma mutation-landscape release in BioTransfer. Five public cBioPortal cohorts are connected with source IDs, retrieval dates and patient/sample denominators. Headline gene frequencies use the MSK 2023 reference cohort (176 unique patients); cohort-specific values remain separate.

The page covers ALK, MYCN, ATRX, TERT, TP53, PHOX2B and BARD1. A zero means no call in the selected mutation profile, while a null means that the source did not support the requested metric.

Scope and current data status

The intended scope includes SNVs, small indels, amplification, copy-number gain, deletion, structural variants, rearrangements and fusions. Mutational signatures are reserved for a later release.

The local snapshot contains five cBioPortal cohorts on hg19. TARGET and MSK include repeat samples, so patient and sample counts are shown separately. Unknown is represented as null, with a status such as unknown, not_assayed or unavailable; it is never converted into zero. The chromosome-wise rollup covers the seven selected genes and loaded representative events, not a genome-wide call set.

Observed data versus interpretation

Observed data

BioTransfer-derived interpretation

A clinically useful comparison of recurrent events, drivers and therapeutic actionability must wait for normalized cohort records, event-specific tested denominators and linked evidence. Somatic observations must remain separate from germline susceptibility evidence, especially for PHOX2B and BARD1.

Cohort and provenance requirements

Each imported cohort must retain its source name, source URL, accession/study ID, source version, retrieval date, licence, assay type, sequencing method, genome build, disease subtype, age group and primary/metastatic status. Each observation must preserve raw, normalized and derived representations.

Frequencies are only valid when numerator and denominator use the same counting unit. Multiple samples from one patient must not inflate a patient-level numerator. Cohorts cannot be pooled until sample overlap, disease definitions and assay coverage are reconciled.

Normalization rules

The importer must retain the source genome build and record any GRCh37/GRCh38 conversion, left normalization, transcript mapping and HGVS representation. SNV/indel, CNV, structural-variant and fusion records must keep event-appropriate representations rather than being forced into one SNV shape.

Limitations in this release

Next data-import step

The next local release should add broader gene coverage, richer copy-number and structural-variant records, clinical-evidence adapters, and a cross-study patient-overlap check before enabling any pooled estimate. The same contract can subsequently support other cancers without changing the page model.

Return to the mutation landscape page