For faster mapping of many sequences against the same reference database
# merge all E. coli reference genomes into one ecoli.fasta file
cat genomes/ecoli_strains/*.fna > ecoli.fasta
# create species genome database for Blast
makeblastdb -in ecoli.fasta -parse_seqids -dbtype nucl
-parse_seqids is required to identify gene/contig IDs in mapping hits
-dbtype nucl specifies the type of sequences: protein 'prot' or nucleotide 'nucl'
# run Blast using the created species DB (database option: -db)
blastn -query genes.fasta -db ecoli.fasta -outfmt 6 -evalue 1e-30
http://www.ncbi.nlm.nih.gov/books/NBK279688/