Retrieve Protein Id And Description From Fasta Header Using Python?

Question

[Martin]

I have looked through all of the tagged Fasta posts, none seem to address this particular issue.

I have the following format for my headers:

>GeneDB|LinJ.35.4080 | organism=Leishmania_infantum | product=ATP-dependent RNA helicase, putative | location=LinJ.35:1596105-1598177(+) | length=690

I would like to learn how to retrieve the "LinJ.35.4080" (let's call this proteinID) as well as the "product=ATP-dependent RNA helicase, putative" (call this proteindescription) fields for all the records in my file and create a two column text file (proteinID, proteindescription).

If known, it would be helpful to remove the "product=" from the field as well.

I have searched through all of the online tutorials (that I could find) related to FASTA, Biopython, Python, etc., and thought (foolishly) my first attempt at this would go a bit smoother.

Thank you in advance for your help and time, this is something that will be used often. I'm hoping I didn't overlook someone else's solution to this.

Answer 1

[Ijessie]

Import regular expression in Python here:

$ python
>>> import re
>>> s = ">GeneDB|LinJ.35.4080 | organism=Leishmania_infantum | product=ATP-dependent RNA helicase, putative | location=LinJ.35:1596105-1598177(+) | length=690"
>>> m = re.search(r'(?m)\>GeneDB\|([^|]*)\s\|.*product=([^|]*)\s\|', s)
>>> if m:
...     pro_id = m.group(1)
...     pro_de = m.group(2)
... 
>>> pro_id, pro_de
('LinJ.35.4080', 'ATP-dependent RNA helicase, putative')

Answer 2

[Neilfws]

No need for Python or any complicated programming. Assuming that all your FASTA headers are of the same form with fields separated by "|" in file myfile.fa, just use grep + awk + sed:

grep "^>" myfile.fa | awk 'BEGIN {FS="|"} {print $2, $4}' | sed 's/product=//'

Result:

LinJ.35.4080   ATP-dependent RNA helicase, putative

[Robert Ernst]

Agree, but if you really want to use python, you can use .split -> http://docs.python.org/library/string.html#string.split