Q. Why is CellWhere wrong?

A. It isn't! (usually*)

CellWhere retrieves a query protein's subcellular localizations from UniProt and/or the Gene Ontology. These resources include several thousand localization terms, that we have grouped into around 50 'CellWhere localizations'.

All of the terms for a given protein are mapped to the CellWhere localizations, which are then ranked according to either:

1. Their frequency (if you select generic localization flavor) or
2. The prioritization score (if you choose a pre-made flavor or upload your own home-made flavor)

In this way, CellWhere selects either the most commonly annotated location, or the location that you have defined as being of the most interest to you.

If you want to see where a protein is generally reported to localize, try using 'Uniprot Only' as the localization source and the generic option as the localization flavor. If you want to highlight locations of special interest and identify any proteins that might be at those locations, then try using 'Uniprot and GO' as the localization source and a pre-made or home-made localization flavor.

* Of course, none of this erases the possibility that we could have a bug in our code or mapping tables, so if you still think CellWhere is wrong, then please PLEASE tell us at: