This track shows variants of the human TP53 tumor-suppressor gene from the UMD TP53 database, a locus-specific variant database maintained by Thierry Soussi and Bernard Leroy. It is a comprehensive, curated resource for TP53 mutations in human cancer, integrating data from tumours, cell lines, germline (Li-Fraumeni syndrome and other hereditary conditions), and non-neoplastic diseases reported in the published literature.
The track is built from the database's variant file, which lists each unique TP53 variant once. For every variant the database provides curated metadata including a pathogenicity classification, the residual transactivation activity of the mutant protein on eight p53 target promoters, functional impact predictions from multiple algorithms (SIFT, PolyPhen-2, MutationAssessor, PROVEAN, Condel, MutPred Splice), and frequencies in tumours, cell lines, somatic samples, and germline cases.
TP53 hot spots such as p.R175H, p.R248Q, p.R248W, p.R273H, p.R273C, and p.R282W are visible as the densely-recurrent positions in the DNA-binding domain.
Each variant is shown as a single coloured box at its genomic position. The colour encodes the database's pathogenicity classification:
| Pathogenic | |
| Likely Pathogenic | |
| Possibly pathogenic | |
| VUS (variant of uncertain significance) | |
| Benign | |
| Unknown |
Hovering over a variant displays its HGVS cDNA name, protein change on the canonical TP53 alpha isoform (NP_000537.3), pathogenicity call, effect class, tumour frequency, total occurrence count, and the curator note on residual transactivation activity. Clicking opens the full details page with the per-promoter activity values, all algorithm-level predictions and scores, the structural domain annotation, links to COSMIC and dbSNP, and the curator's final comment.
The track configuration page supports filters on:
Variants can be searched from the position box by HGVS cDNA name (e.g. c.524G>A) or by canonical protein change (e.g. p.R175H).
The track is rebuilt automatically when a new release of the UMD TP53 database is detected at p53.fr.
The raw data can be explored interactively with the
Table Browser or the
Data Integrator. The underlying
bigBed file may be downloaded directly
from /gbdb/$db/bbi/umdTp53/umdTp53.bb, and the data can be
queried programmatically through the
REST API via the track name
umdTp53.
The variants file from p53.fr (UMD_variants_US.tsv, one row per unique
TP53 variant) was reformatted to
bigBed. For hg19, genomic coordinates
were taken directly from the variants file's HG19_Start and
HG19_End columns. The variants file does not ship hg38 coordinates, so
for hg38 the coordinates were joined from the companion mutations file
(UMD_mutations_US.tsv), which contains both hg19 and hg38 positions for
every variant.
The curator's residual-activity comment column derives from the eight-promoter transactivation assay of Kato et al., 2003: for missense variants, the median of the eight per-promoter activities (expressed as % of wild-type) is mapped to one of Fully active, Hyper active, Partial activity, No activity, or No data. For nonsense, frameshift, splice-site, and similar variants where the assay does not directly apply, the column carries an explanatory note (e.g. The activity of truncated p53 is assumed to be nil). The eight raw per-promoter values are available on the details page.
Thanks to Thierry Soussi, Bernard Leroy, and the UMD TP53 database curators for maintaining the resource and making it available for download.
Leroy B, Anderson M, Soussi T. TP53 mutations in human cancer: database reassessment and prospects for the next decade. Hum Mutat. 2014 Jun;35(6):672-88. PMID: 24665023
Kato S, Han SY, Liu W, Otsuka K, Shibata H, Kanamaru R, Ishioka C. Understanding the function-structure and function-mutation relationships of p53 tumor suppressor protein by high-resolution missense mutation analysis. Proc Natl Acad Sci U S A. 2003 Jul 8;100(14):8424-9. PMID: 12826609; PMC: PMC166245