Protein sequence from EC number and species

Question

[miquelduranfrigola]

Hi all,

From BRENDA database I got a list of ~20,000 pairs of (EC number, Organism) that are of my interest. I need the protein sequence corresponding to each of these enzymes. In principle, BRENDA provides protein sequences (you can fetch them through the SOAP server), although the coverage is pretty low (I could only obtain a sequence for ~8,000 of the pairs).

Is there a better way to obtain enzyme sequences only from EC number and Organism?

Many thanks!

Miquel

Answer 1

[Brett]

The PIR database is your best bet.

You can select function based upon various criteria (e.g. EC number, GO ID, Kegg Pathway, Enzyme name etc) and you can select organism in a separate field (tax ID, organism name etc).

This is probably the most powerful solution (array of 100+ criteria you can define your selection) and at the same time the most usable.

Sequences can be downloaded directly (I often do a few thousand at a time) as a fasta.

Answer 2

[battle_toad]

You can try to retrieve sequences from uniprot using rest interface.

Example: (ec: 1.1.1.1 , organisms: Homo sapiens)

http://www.uniprot.org/uniprot/?query=ec:1.1.1.1+AND+organism:"Homo sapiens"&format=fasta

Full information abut syntax you can find here