Last modified July 6, 2015
This subsection of the ‘Subcellular location’ section describes the subcellular compartment where each non-membrane region of a membrane-spanning protein is found.
Topological domains are annotated according to a controlled vocabulary, the elements of which are listed below.
Common topological domains
- Chloroplast intermembrane
- Lumenal, thylakoid
- Lumenal, vesicle
- Mitochondrial intermembrane
- Mitochondrial matrix
- Peroxisomal matrix
- Perinuclear space
- Virion surface
Rare topological domains
- Lumenal, melanosome
- Mother cell cytoplasmic
- Forespore intermembrane space
- In membrane
- Pore forming
- Exoplasmic loop
Note that in each case the subcellular location of the protein as a whole is defined in the ‘Subcellular location’ subsection.
We annotate both experimentally determined and predicted topological domains.
1. Annotation of experimentally proven topological domains:
Even when supporting experimental evidence is available for the assignment of a particular region of a protein to a particular cellular compartment, the topological domains of a protein are generally annotated using the evidence ‘Sequence analysis’ due to the inherent difficulties in determining the precise boundaries of the neighbouring transmembrane regions.
However when the experimental technique used allows the unambiguous assignment of the transmembrane boundary to a particular position (X-ray crystallography, etc.), the ‘Sequence analysis’ qualifier is not added in the topological domain annotation. In this case, the positions of the topological domains can be propagated ‘By similarity’.
2. Annotation of predicted topological domains:
We also annotate topological domains based on the predictions provided by the program TMHMM, which returns a probability that non-transmembrane regions of a protein are in a cytoplasmic compartment (specified as ‘in’) or a non-cytoplasmic compartment (specified as ‘out’). When TMHMM returns a high confidence topology prediction for a protein of known subcellular location, the notional ‘in’ and ‘out’ predictions from TMHMM are specified using a controlled vocabulary as described above. For instance, we may specify ‘in’ as ‘cytoplasmic’ and ‘out’ as ‘periplasmic’.
When nothing is known about the subcellular location of the protein, we annotate the topological domains based on TMHMM predictions using the default values of ‘cytoplasmic’ and ‘extracellular’. In all cases, predicted topological domains are annotated with ‘Sequence analysis’.