I had a protein Refseq ID and I PSI-BLASTed this sequence against Refseq database. We all know that the Refseq is a Reference sequence database and it shouldn't have redundancy. After BLASTing my sequence, at first iteration I got 1000 hits and among them there were a lot of redundant sequences!
My sequence had 241 amino acids and I found a lot of sequences with 100% identity, 100% cover and 0 E-value exactly the same as my sequence but with different IDs. All of these IDs were from RefSeq! In other iterations and after adjusting format options, I got this redundancy with other sequences from other species. My sequence is related to a chain of a multichain protein (E.coli fumarate reductase)
For example, when using NP_418578 as a query, I found WP_078165098.1, WP_064226696.1, WP_062863447.1, WP_001401474.1 and other that were identical.
I want to know what is wrong with Refseq. Is it really a Reference sequence database? If it is, where does this redundancy come from and why?