Blast databases available locally

Many pipelines involving annotation/assembly comparison involve Blast ( Several Blast versions are available as modules, for example:

  • blast/2.7.1+, etc. : the Blast+ suites (blastp, tblastn, etc.), recommended
  • blast/2.2.26, etc. : 'legacy' Blast (blastall, megablast, etc)

Use module spider blast to see available versions. As for all bioinformatics tools at Uppmax, module load bioinfo-tools is required before the blast modules are available.

Uppmax maintains local copies of many Blast databases, including those available at NCBI ( as well as several UniProt databases ( Note that:

  • The local copies are found at /sw/data/blast_databases
  • New versions are installed the first day of each month at 00.01 from local copies updated the 28th of the previous month beginning at 00.01
  • When new versions are installed, the directory containing the previous versions is renamed to blast_databases_old
  • blast_databases_old is deleted the second data of each month at 00.01

The databases available are:

Name Type Source Notes
cdd_delta protein NCBI Conserved domain database for use with delta-blast
env_nr protein NCBI Protein sequences for metagenomes (EXCLUDED from nr)
env_nt nucleotide NCBI Nucleotide sequences for metagenomes
human_genomic nucleotide NCBI Human RefSeq chromosome records
nr protein NCBI Non-redundant protein sequences
nt nucleotide NCBI Partially non-redundant nucleotide sequences
other_genomic nucleotide NCBI RefSeq chromosome records for non-human organisms
pataa protein NCBI Sequences for protein structures from patents (EXCLUDED from nr)
patnt nucleotide NCBI Nucleotide sequences for pataa protein structures from patents
pdbaa protein NCBI Sequences for protein structures from Protein Data Bank
pdbnt nucleitide NCBI Nucleotide sequences for pdbaa protein structures from Protein Data Bank
refseq_genomic nucleotide NCBI NCBI genomic reference sequences
refseq_protein protein NCBI NCBI protein reference sequences
refseq_rna nucleotide NCBI NCBI Transcript reference sequences
swissprot protein NCBI Swiss-Prot sequence database
uniprot_sprot protein UniProt Swiss-Prot high quality manually annotated and non-redundant protein sequence database
uniprot_trembl protein UniProt TrEMBL high quality but unreviewed protein sequence database
uniprot_sptrembl protein uniprot_sprot and uniprot_trembl combined
uniprot_all protein alias for uniprot_sptrembl
uniprot_all.fasta protein alias for uniprot_sptrembl
uniprot_sprot_varsplic protein UniProt UniProt canonical and isoform sequences (see link)
uniprot_uniref50 protein UniProt Clustered sets of 50%-similar protein sequences (see link)
uniprot_uniref90 protein UniProt Clustered sets of 90%-similar protein sequences (see link)
uniprot_uniref100 protein UniProt Clustered sets of identical protein sequences (see link)
UniVec nucleotide UniVec Sequences commonly attached to cDNA/genomic DNA during the cloning process
UniVec_Core nucleotide UniVec A subset of UniVec chosen to minimise false positives

Additionally, taxdb.btd and taxdb.bti are downloaded, which provide additional taxonomy information for these databases. Local copies of the NCBI Taxonomy databases are also available; further details are available on a separate page.

For UniVec and UniVec_Core, Fasta-format files containing the vector sequences are also available with the given names (e.g., /sw/data/blast_databases/UniVec), alongside the Blast-format databases built from the same Fasta files.

The exact times all databases were updated are provided by database.timestamp files located in the directory

Databases are available automatically after loading any blast module

When any of the blast modules is loaded, the BLASTDB environment variable is set to the location of the local database copies (/sw/data/blast_databases). The various Blast tools can use this variable to find the locations of databases, so that only the name needs to be specified.

module load bioinfo-tools blast/2.7.1+
blastp -db nr -query input.fasta

After loading the blast/2.7.1+ module, specifying blastp -db nr results in blastp searching the local copy of nr, because the BLASTDB environment variable is set when the module is loaded. Similarly, each of these would result in searching the local copy of the given database:

blastp -db pdbaa ...
blastp -db uniprot_sprot ...
blastp -db uniprot_uniref90 ...
blastn -db nt ...
blastn -db refseq_genomic ...

WGS, TSA and SRA sequence databases are not included

The NCBI Whole-Genome Shotgun is no longer downloaded locally. NCBI provides special versions of Blast and other tools that can be used to search the remote versions of WGS, the Transcriptome Shotgun Assembly, and the Sequence Read Archive.

These special blast versions and other tools are part of NCBI's SRA Tools, which is available at Uppmax as the sratools module. We have also include auxiliary NCBI scripts in the sratools module to convert taxonomic IDs to WGS and TSA identifiers.