A database of mammalian mitochondrial localisation evidence, phenotypes and diseases
Experimental data is available for the following species:
GFP tagging provides excellent evidence for mitochondrial localisation and is considered to be more reliable than mass spectrometry data. Due to experimental technical difficulties most of these studies are limited to yeast.
As mass spectrometry data can include false positives from the incomplete removal of non-mitochondrial proteins in the sample used, we have also recorded the various methods for the purification, separation and identification used in each study, to provide a full provenance for each entry. Proteins with minimal mass-spec evidence are not necessarily non-mitochodrial proteins that have been identified due to contamination. If a protein is low abundance, tissue specific or expression specific this may result in it being missed by a majority of studies. In addition many mitochondrial proteins are membrane bound and these are notoriously difficult to identify by mass spectrometry as most purification procedures are biased towards extracting soluble proteins.
For each mass spectometry protein record the following information is extracted and loaded into MitoMiner where available:
Several of these studies conducted multiple experiments on several tissues and MitoMiner records entries from the same study but performed on different tissues separately.
MitoMiner includes homology information from Ensembl Compara to allow mitochondrial evidence to be shared between different species for the same homologous gene.
We also include immunofluorescent staining localisation data from the Human Protein Atlas. We include the original gene identifier, main location, other location, expression type and reliability. The 'other locations' were classed by the Human Protein Atlas as those with a markedly lower staining intensity than the main location, or those that were only observed in a subset of the cell lines. Data based on a single antibody is reported as expression type - staining. If the data is based on more than one antibody it is reported as expression type - APE. Reliability is classed as whether the antibodies staining agrees if more than one is used and/or if the localisation agrees with UniProt. A reliability of strong, medium and supportive are considered to be good evidence for that localisation.
These links will take you to the Query Builder pages which allow you to make new queries and create template queries of your own