News:Ncbi Releases Entrez Direct, The Entrez Utilities On The Unix Command Line

Question

[Pierre Lindenbaum]

http://www.ncbi.nlm.nih.gov/news/02-06-2014-entrez-direct-released/

NCBI has just released Entrez Direct, a new software suite that enables users to use the UNIX command line to directly access NCBI databases, as well as to parse and format the data to create customized downloads.

Retrieve a set of PubMed abstracts

(...) (...)

[Medhat]

O so happy to here that :D

Documentation is here:

Documentation

FTP

[BioApps]

I should understand that it works on Linux also. Right?

Answer 1

[Giovanni M Dall'Olio]

I have been trying these utilities in the last few days. They work nice, but the documentation is very obscure! There is not even an --help flag implemented, and you have to go back to the web page every time.

Anyway, here are a few examples, not covered in the in the documentation:

Given a Gene ID, download the aminoacid sequences of the corresponding Proteins, keeping only the reviewed entries (e.g. no putative, predicted sequences):

esearch -db gene -query "1234[id]" | elink -target protein | efilter -query "REVIEWED[FILTER]"| efetch -format fasta

Given a file containing a list of Gene IDs (one per line), download all the entries in tabular format:

esearch -db gene -query $(paste -s -d ','  mygenes.ids) | efetch -format tabular > mygenes.details.txt

Please add more examples!

Answer 2

[Istvan Albert]

First thought: does this mean that my home grown, undocumented, hacky curl based shell script is obsolete?

Second thought: What took them so long?

[5heikki]

I'm just happy they finally replaced eutils with something more user-friendly..

[pld]

I couldn't agree more, e-utils was awful. I'm still not a total fan of relying on entrez queries to obtain larger amounts of sequence data. I've not had a good experience with getting the accuracy I need, there's always something that I don't want in there.