Subcellular localisation is determined using both N- and C-terminal GFP fusions for every ORF. This maximises the possibility of correctly ascertaining the localisation, since the presence of GFP may mask targeting signals that may be present at one end of the native protein. Similar localisations with both N- and C-terminal constructs also provides a higher degree of reliability of the result. For those ORFs where the two fusions do not give a similar localisation pattern, we have a series of other criteria, including bioinformatic predictions, which we also consider in order to help us make our decision. These are explained in greater detail on the ‘FAQs’ page. Occasionally proteins localise to more than one type of structure. In these cases we still annotate a ‘final’ localisation category based on the results we have.
Initial localisations are all made using transient transfection in living cells. This has the advantage that it is rapid, allows us to monitor the cells at different time points, and permits the identification of motile structures. All the images shown on this website are of living cells, including the examples shown below. At the end of the live cell imaging, the cells can still be fixed and colocalisation experiments made.