Welcome to the UMD TP53 mutation database

Current release: June 2012 (2012_R1)

This new release of the UMD TP53 mutation database has been updated, improved and provides several novel and unique features.
Three objectives were taken into account in the development of this new release of the TP53 database

Firstly, we have curated the database to tag and/or remove publications that are notoriously (or less notoriously) artefactual. Curation was performed using both PCA (Principal Component Analysis) and K-mean clustering (see the curation page for more info). Combined with statistical analysis, it has been possible to tag publications that report suspicious set of mutations. This type of analysis is far superior to and more objective than manual curating. The UMD TP53 is the only TP53 database that has been actively curated using multiple statistical analyses.

Edlund K, Larsson O, Ameur A, Bunikis I, Gyllensten U, Leroy B, Sundstrom M, Micke P, Botling J, Soussi T (2012) Data-driven unbiased curation of the TP53 tumor suppressor gene mutation database and validation by ultradeep sequencing of human tumors. Proc Natl Acad Sci U S A 109: 9551-9556.
Download the pdf

For greater flexibility, the UMDTP53 database is available in two versions, curated or uncurated. In both sets, each publication is associated with a confidence tag that allows the user to work with a specific dataset (see the curation page and the read me file for more info).

Secondly, we provide an accurate and easily accessible database with data that can be conveniently explored, analysed, filtered or retrieved for accurate evaluation.

Thirdly, we povide novel information allowing the user to perform statistical analysis on TP53 mutants. In particular, this revised version includes the various raw parameters used to score TP53 mutations.

Analytical parameters

Name of the field in the database

Usage

*Frequency of each mutation in the database

Number of records

Distinguish hot spot mutations and infrequent mutants

*Frequency of each mutation in a single publication

Publication occurance

A large number of infrequent mutations in a single publication is highly suspicious

Number of other mutations found in the same tumours

Multiple_mut

Values of three and more are highly suspicious and should be very rare

Number of unidentified mutations found in the same tumours

Unidentified mutation

 

Specific tag to distinguish various types of tandem mutations

Tandem class

Syn mutations in tandem mutations must not be considered in the same way as those observed as single events

Specific tag to distinguish tumours with two mutations in the same allele (DMS), tumours with 2 mutations in different alleles (DMD) or with an unknown distribution (DMU)

Complexity

If the mutations are on different alleles, then two different TP53 mutants are expressed and must be considered as individual mutants for any analysis.

*The ratio of these two values (Freq_Publi / Freq_Database) for mutants found several times in the database is highly suggestive of suspicious mutations. A value of 1 indicates that the mutation has been described in only one publication.

Fourthly, for each TP53 mutant, we provide specific information regarding its frequency in the database, its proximity to a splice site or its association with dubious studies.

Fitfhtly, each publication is associated with confidence tags that allow the user to filter the database using his own criteria.

Users can find more info on the curation page.

any question or specific enquiries: p53@free.fr