This gene, a member of the cadherin superfamily, encodes one of the receptor tyrosine kinases, which are cell-surface molecules that transduce signals for cell growth and differentiation. This gene plays a crucial role in neural crest development, and it can undergo oncogenic activation in vivo and in vitro by cytogenetic rearrangement. Mutations in this gene are associated with the disorders multiple endocrine neoplasia, type IIA, multiple endocrine neoplasia, type IIB, Hirschsprung disease, and medullary thyroid carcinoma. Two transcript variants encoding different isoforms have been found for this gene. Additional transcript variants have been described but their biological validity has not been confirmed.
Tyr: tyrosine protein kinase.
(Log odds ratio: 0.628591)
The JMOL viwer can load structure PDB files associated to that UniProt
protein, as well as models provided by Interactome3D. The SVG image and the
sequence scroller bellow correspond to the Ensembl protein associated with
that UniProt: ENSP00000347942
In order to adress all inconsistencies between the sequences in UniProt, Ensembl,
and UniProt, coordinate maps are built using Smith-Waterman pairwise alignment.
The following information is extracted from Structure-PPi, a workflow for
the annotation of protein residues. It finds features overlapping the
variant or overlapping any residue that is in close physical proximity
to the variant: 5 angstroms spatial distance or adjacent in the sequence
if no PDB covers that area.
It may also report protein protein interfaces that might be affected by
the variant, i.e. when they are at a distance of 8 angstroms from
residues from a partner protein.
The structure PDB files used for these calculations are all those
associated with the UniProt protein in UniProt and Interactome3D, which
extends this set with models of single proteins and protein-protein
interaction complexes. Pairwise alignment is used to resolve any sequence
The databases examined are UniProt, InterPro, COSMIC and Appris. The
COSMIC database reports variants affecting that residue, not necessarily
the same amino-acid change. Appris agglutinates features from FireDB,
which are catalytic sites, and ligand-binding sites, and others such as
trans-membrane domains, signal peptides, and ligand binding sites.
A module for the annotation of cancer-related single-nucleotide variants at protein-protein interfaces.
Miguel Vazquez, Alfonso Valencia, Tirso Pons.2015.Bioinformatics.
KinMut Random Forest is a method specific for prediction of the pathogenicity
of variants affecting the protein kinase superfamily. It characterizes the
variants at the gene, domain and residue level with a combination of general
and kinase-specific features. The pathogenicity of variants is evaluated by a
random forest algorithm. Predictions are accompanied with a reliability score.
Our approach outperforms available methods in a cross-validation experiment on
the 3689 kinase variants in Uniprot. (Accuracy: 88.45%, Precision: 81.62%,
recall: 75.22%, f-score: 78.29% and MCC: 0.68)
KinMut Random Forest
disease if score >0, neutral if score <0
Group log-odds ratio
GO log-odds ratio
Affected protein domains
Log odds ratio
Other predictors (dbNSFP)
Precomputed prediction not found on dbNSFP for ENSP00000347942:A883F
These literature co-mentions have been extracted automatically from the literature with
Although it has proved to be a powerful guide to gather contextual
information, manual inspection of these results might be required. Refer to
Hoffmann and Valencia, 2004
for further details.
Loading. Please Wait.
CNIO. Centro Nacional de Investigaciones Oncologicas (Spanish National Cancer Research Center)
Structural Computational Biology Group C/ Melchor Fernandez Almagro, 3, E-28029 Madrid