EBI UniParc vs NCBI NR protein databases

Question

[cdsouthan]

UniParc https://www.ebi.ac.uk/uniparc/ is at 632,168,010 sequences and NCBI NR is 648,450839

Does anyone know if the diff between these includes useful sequences in one but not t'other?

[GenoMax]

"useful" is probably relative to the user. The differences may be simply be due to time/processes the two databases use. With time missing sequences may make it into either (unless a source is not being looked at at all). UniParc lists their sources but not NCBI as far as I see.

[cdsouthan]

Ta for reply. The NCBI result return says ":All non-redundant GenBank CDS translations+PDB+SwissProt+PIR+PRF excluding environmental samples from WGS projects"
Patent proteins not specified (as they are for UniParc) but I guess are also in there. W.b.g to know where those +10K in NR originated from. As ever we have to live with the transatlantic two-stop-shops without knowing precisely what is behind the counter.....