Anyone can submit sequences to Uniprot (and it its true for almost any biological database), so eventually you will have this kind of redundancy. Uniprot usually try to reduce the redundancy, but in cases were there is difference in the organism (eg: different genus, species, sub-species, strain, lineage, isolate, serovar, pathovar, serogroup, chemiovar ....) it might not be applied, or even worst, inviabilize future evolutionary studies.

The name "POU Domain" is an estrutural annotation, not the most correct name of the protein, in fact. Many proteins in the same organism may contain these domain and be encoded by different genes; be from the same paralogous group or family of genes; or be isoforms of the same genes. In all of these cases, or many others, it won't count as a redundancy at all. The fact that these entries also have the same gene also doesn't necessary mean that they are de facto encoded by the same locus, as genes might be eventually duplicated in the genome, and the responsible for the annotation of the protein or the genome is usually who gives the name that is displayed, not the database admin (usually annotation is made by automated tools).

So, for many reasons, Uniprot will keep these cases of aparently-redundant records, because specially because it is very probable that the records are not in fact redundant.

Hi Wasim,

one gene can lead to multiple transcribes, leading to multiple proteins..

fred