How to fetch a nucleotide sequence of a certain CDS from linux terminal by using GenBank accession number.

Question

[arriyaz.nstu]

Let's say, I have two accession numbers AP017458 & GQ994935 of a virus's GenBank file. I want to download the nucleotide sequence (CDS) of ORF57 from these accession from Linux terminal. As per as I know efetch, esearch, etc command line code can be used to download sequence directly from the terminal. But, I need only the coding sequence of a specific ORF.

How can I download the coding sequence of a specific ORF by using accession number (all together or, at least one accession per each run)?

Answer 1

[GenoMax]

Using EntrezDirect (sequence truncated for space) :

$ esearch -db nuccore -query "AP017458" | elink -target protein | esummary | xtract -pattern DocumentSummary -if Title -contains "ORF57" -element Extra | awk -F "|" '{print $4}' | xargs -n 1 sh -c 'efetch -db protein -id $0 -format fasta_cds_na' 

>lcl|AP017458.1_cds_BAV17910.1_1 [protein=ORF57] [protein_id=BAV17910.1] [location=join(82084..82131,82240..83559)] [gbkey=CDS]
ATGGTACAAGCAATGATAGACATGGACATTATGAAGGGCATCCTAGAGGACTCTGTGTCCTCCTCTGAGT
TTGACGAATCGAGGGACGACGAGACGGACGCACCGACACTGGAAGACGAGCAATTGTCCGAACCCGCCGA
GCCTCCGGCAGACGAGCGCATGCGTGGTACCCAGTCGGCCCAGGGAATCCCACCCCCCCTGGGCCGCATC

There are two entries for the other accession

$ esearch -db nuccore -query "GQ994935" | elink -target protein | esummary | xtract -pattern DocumentSummary -if Title -contains "ORF57" -element Extra | awk -F "|" '{print $4}' | xargs -n 1 sh -c 'efetch -db protein -id $0 -format fasta_cds_na' 

>lcl|GQ994935.1_cds_ACY00456.1_1 [protein=ORF57] [protein_id=ACY00456.1] [location=join(81886..81934,82043..83361)] [gbkey=CDS]
ATGGTACAAGCAATGATAGACATGGACATTATGAAGGGCATCCTAGAGGACTCTGTGTCCTCCTCTGAGT
TTGACGAATCGAGGGACGACGAGACGGACGCACCGACACTGGAAGACGAGCAATTGTCCGAACCCGCCGA

>lcl|KF588566.1_cds_AKE33094.1_1 [gene=ORF57] [protein=ORF57] [protein_id=AKE33094.1] [location=join(81886..81934,82043..83361)] [gbkey=CDS]
ATGGTACAAGCAATGATAGACATGGACATTATGAAGGGCATCCTAGAGGACTCTGTGTCCTCCTCTGAGT
TTGACGAATCGAGGGACGACGAGACGGACGCACCGACACTGGAAGACGAGCAATTGTCCGAACCCGCCGA
GCCTCCGGCAGACGAGCGCATCCGTGGTACCCAGTCGGCCCAGGGAATCCCACCCCCCCTGGGCCGCATC

[arriyaz.nstu]

Hi, your code is just amazing. Thank you very much. Is it possible to download the CDS from more than one accession together in a single code? Becuase I have a very long list of accession numbers and I've to download CDS of ORF57 for all of them.

[GenoMax]

Absolutely. Use epost method :

 $ epost -db nuccore -input id_file | elink -target protein | esummary | xtract -pattern DocumentSummary -if Title -contains "ORF57" -element Extra | awk -F "|" '{print $4}' | xargs -n 1 sh -c 'efetch -db protein -id $0 -format fasta_cds_na'

id_file should contain one accession per line.

[arriyaz.nstu]

You made my day. Thank you very much.

[GenoMax]

Please accept the answer (green check mark) then.