Hello, Do the nr.gz database still corrupted? I've download it twice but i still get the same error. any help?

du -sh nr.gz 
187G    nr.gz
(base) blackboy@LAPTOP-814NSMMO:~/db$ gunzip -c nr.gz \
 | sed '/^>/s/ .*//' \
 | diamond makedb \
      --taxonmap prot.accession2taxid.FULL.gz \
      --taxonnames names.dmp \
      --taxonnodes nodes.dmp \
      --db ncbi-nr.taxonomy.dmnd
diamond v2.1.8.162 (C) Max Planck Society for the Advancement of Science, Benjamin Buchfink, University of Tuebingen
Documentation, support and updates available at <http://www.diamondsearch.org>
Please cite: <http://dx.doi.org/10.1038/s41592-021-01101-x> Nature Methods (2021)

#CPU threads: 4
Scoring parameters: (Matrix=BLOSUM62 Lambda=0.267 K=0.041 Penalties=11/1)
Input file parameter (--in) is missing. Input will be read from stdin.
Opening the database file...  [0.049s]
Loading sequences...
gzip: nr.gz: invalid compressed data--format violated
Error: Error reading input stream at line 206466: Invalid character ( ) in sequence

I don't know. It's so large that the risk that the download fails is high. When I download such large files I use

wget -c

that can take up an interrupted download again (depends on the remote). RefSeq is quite a bit smaller and good enough for taxonomy in our pipeline.

Ok, could you please advise me another way (db) to work with?

The NCBI databases in combination with the Diamond tool is the best option for a general taxonomy, i.e. covering everything. There is not, I believe, a fasta file available for the slightly smaller RefSeq, however. One can download the blast database for RefSeq, then use

blastdbcmd

to extract a fasta file from it. Since that's more complicated than using the

nr.gz

file I would start by trying to download the file again, using

-c

to make sure the download is retried if it fails.