How to convert a database from protein to nucleotide

Question

[AJTrunkskun94]

Hi! I'm fairly new at UNIX and Bioinformatical work. I am taking a class now that is focusing on using UNIX and BLAST on nucleotide sequences. When the databases were made, we have been using blastn, constantly. I wanted to apply some of what I learned to my own research and realized that my databases are now protein databases, even though the fasta file is ALL nucleotides. Is there a way to convert the database from protein to nucleotide?

[WouterDeCoster]

Due to redundancy in codon usage for multiple amino acids a protein sequence does not uniquely identify a nucleotide sequence.

[GenoMax]

If you only have the index files (i.e. no fasta protein sequence) then you would need to use blastdbcmd utility to first recover the fasta sequence.

Then you can use back-translation tools like backtranseq from EMBOSS (http://www.ebi.ac.uk/Tools/st/). Ideally if you know what genome those proteins are from then you could go and get the DNA sequence from source instead.

[AJTrunkskun94]

The genome is from Listeria monocytogenes, and I have the fasta file for a contig of this genome, but for some reason it recognizes it as a protein and not nucleotide file. Even though the first couple lines are: AGATTCCTTGCGTCAAATTGACTTCGCTAGCAATTAAATTACTAGTTTGTTTTGTTGAAAACAGCTTTCT GTTTTCTGCCCTGCGATTACCAGTGAGACTTTACGTCTCATTGCTTTTCGTCTTCTTCTTTGTTCAGTTT TCAAAGGTCAGTTGCTTTGTTAACGCAACTTTTAAATCTTACCATAAAGTTGAAATCACGTCAACAACTA

[GenoMax]

That is definitely not protein and if you started at the top of the file it is not in fasta format either.

but for some reason it recognizes it as a protein and not nucleotide file.

What is doing that? Blast? Have you created indexes for this dataset already?

[AJTrunkskun94]

So when I made the databases by makeblastdb -dbtype prot -in LM_R8_5081_contig10.fasta -out LM_R8_5081_contig10 -parse_seqids Then the report after says that it is a protein: New DB title: LM_R8_5081_contig10.fasta Sequence type: Protein Deleted existing Protein BLAST database named /home/ajt3/Listeria_Work/LM_R8_5081_contig10_test Keep Linkouts: T Keep MBits: T Maximum file size: 1000000000B

[AJTrunkskun94]

Nope, I'm dumb, I realized what it is...

[GenoMax]

Glad you figured the problem out yourself :)

[st.ph.n]

realized that my databases are now protein databases, even though the fasta file is ALL nucleotides.

It sounds like the database you have, is a nucleotide fasta, but when you made the db, it was made as a protein database, so the extensions are incorrect for the indexed files.

Do your db files end with the following?

.phr, .pin, .psq

remake your database with this command:

makeblastdb -in yourfastfile.fasta -dbtype nucl

and your db will be made with the correct extensions for blastn. Your files will not end with

.nhr, .nin, .nsq