Strange "Error: OID not found" in NCBI's blastdbcmd

Question

[juanma_lace]

I'm getting a OID not found when executing blastdbcmd. My file is as follows:

cat data/4D/4ds_newbler2.6_shotgun_lmp_scaffolds.fasta.splitted  | grep scaffold00001_37_1_21
>scaffold00001_37_1_21

And I run

makeblastdb -dbtype 'nucl' -in data/4D/4ds_newbler2.6_shotgun_lmp_scaffolds.fasta.splitted -out data/4D/blastdb-parseqid/4ds_splitted -parse_seqids

Building a new DB, current time: 12/18/2014 14:12:08
New DB name:   data/4D/blastdb-parseqid/4ds_splitted
New DB title:  data/4D/4ds_newbler2.6_shotgun_lmp_scaffolds.fasta.splitted
Sequence type: Nucleotide
Deleted existing BLAST database with identical name.
Keep Linkouts: T
Keep MBits: T
Maximum file size: 1000000000B
Adding sequences from FASTA; added 307 sequences in 0.021672 seconds.

And then

blastdbcmd -db data/4D/blastdb-parseqid/4ds_splitted -entry scaffold00001_37_1_21 -outfmt %f -out 1_scaffold00001_37_1_21.fsa

Error: scaffold00001_37_1_21: OID not found
BLAST query/options error: Entry not found in BLAST database
Please refer to the BLAST+ user manual.

Any ideas?

Thank you in advance

EDIT:

If I take a look to the output of blastdbcmd with entry "all" I cannot find the original sequence, why is it missing?

blastdbcmd -db data/4D/blastdb-parseqid/4ds_splitted -entry all -outfmt %f -out 1_scaffold00001_37_1_21.fsa**
cat 1_scaffold00001_37_1_21.fsa | grep 0001_

nothing.

[Pierre Lindenbaum]

Are you sure about -parse_seqids?

Answer 1

[juanma_lace]

Deleting previous generated databases with makeblastdb solve the issue. I was getting no error, but the database was not being generated again.

Answer 2

[Siva]

If you want to create BLAST databases with an option to retrieve individual sequences using sequence ID, you need to have the sequence ID in a specific format that is described in the BLAST user manual.