convert DNA FASTA to protein FASTA format (faa)

Question

[nkuo]

Evening.

I have de novo assembled whole genome DNA sequences and an embl annotation file that contains CDS regions for multiple genes. I am wondering if there are any tools that can first, parse the embl annotation file to obtain the CDS positions. Then use these CDS genome positions extracted from the embl file to convert each DNA fasta file into multifasta files containing the translated amino acid sequence (.faa) for every CDS range given in the embl file for every isolate.

Thanks

[Matteo Schiavinato]

It's the kind of tool that bioinformaticians usually brew theirselves. I am not aware of tools doing it, but in principle it's quite easy to achieve as long as you have a genetic code in the form of a dictionary / hash.

E.G. for python:

TRANSLATE = {"UUU":"F", "UUC":"F", "UUA":"L", "UUG":"L",
    "UCU":"S", "UCC":"S", "UCA":"S", "UCG":"S",
    "UAU":"Y", "UAC":"Y", "UAA":"STOP", "UAG":"STOP",
    "UGU":"C", "UGC":"C", "UGA":"STOP", "UGG":"W",
    "CUU":"L", "CUC":"L", "CUA":"L", "CUG":"L",
    "CCU":"P", "CCC":"P", "CCA":"P", "CCG":"P",
    "CAU":"H", "CAC":"H", "CAA":"Q", "CAG":"Q",
    "CGU":"R", "CGC":"R", "CGA":"R", "CGG":"R",
    "AUU":"I", "AUC":"I", "AUA":"I", "AUG":"M",
    "ACU":"T", "ACC":"T", "ACA":"T", "ACG":"T",
    "AAU":"N", "AAC":"N", "AAA":"K", "AAG":"K",
    "AGU":"S", "AGC":"S", "AGA":"R", "AGG":"R",
    "GUU":"V", "GUC":"V", "GUA":"V", "GUG":"V",
    "GCU":"A", "GCC":"A", "GCA":"A", "GCG":"A",
    "GAU":"D", "GAC":"D", "GAA":"E", "GAG":"E",
    "GGU":"G", "GGC":"G", "GGA":"G", "GGG":"G",}

[Joe]

If you already have the CDS multifasta file, translating it is fairly trivial. Do you already have this or need to generate it from the EMBL file?

[nkuo]

No, we do not have CDS multifasta file yet, we need to generate from EMBL file

Answer 1

[Joe]

Something like this? Extracting All Cds From A Embl File

Biopython and other parsers should be able to handle EMBL format. It'll store all the CDS features in a seqrecord once you parse it in with (probably) SeqIO. You can then just get it to spit out all the CDS protein features.

I answered basically this task before here A: How do can I use Biopython and SeqIO to parse out multiple genes from several NC (see the # genbank2fasta.py script). You could maybe even consider first converting your EMBL file to a Genbank and then all the scripts in that second thread would work,