#!/usr/bin/env python
# -*- coding: utf-8 -*-
'''Provides a class for downloading sequences from genbank.
'''
import re
import os
import urllib2
import warnings
from urlparse import urlparse
from urlparse import urlunparse
from Bio import SeqIO
from Bio import Entrez
from Bio.Alphabet.IUPAC import IUPACAmbiguousDNA
from Bio.GenBank import RecordParser
from Bio.SeqUtils.CheckSum import seguid
from pydna.dsdna import read
[docs]class Genbank():
'''Class to facilitate download from genbank.
Parameters
----------
users_email : string
Has to be a valid email address. You should always tell
Genbanks who you are, so that they can contact you.
proxy : string, optional
String containing a proxy url:
"proxy = "http://umiho.proxy.com:3128"
tool : string, optional
Default is "pydna". This is to tell Genbank which tool you are
using.
Examples
--------
>>> import pydna
>>> gb=pydna.Genbank("me@mail.se", proxy = "http://proxy.com:3128") #doctest: +SKIP
>>> rec = gb.nucleotide("L09137") # <- pUC19 from genbank #doctest: +SKIP
>>> print len(rec) #doctest: +SKIP
2686 #doctest: +SKIP
'''
def __init__(self, users_email, proxy = None, tool="pydna"):
if not re.match("[A-Z0-9._%+-]+@[A-Z0-9.-]+\.[A-Z]{2,4}",users_email,re.IGNORECASE):
raise(ValueError("Not a valid email address!"))
self.email=users_email #Always tell NCBI who you are
if proxy:
parsed = urlparse(proxy)
scheme = parsed.scheme
hostname = parsed.hostname
test = urlunparse((scheme, hostname,'','','','',))
try:
response=urllib2.urlopen(test, timeout=1)
except urllib2.URLError as err:
warnings.warn("could not contact proxy server")
print { scheme : parsed.geturl() }
self.proxy = urllib2.ProxyHandler({ scheme : parsed.geturl() })
else:
proxy_handler = urllib2.ProxyHandler({})
opener = urllib2.build_opener(proxy_handler)
urllib2.install_opener(opener)
#os.environ['http_proxy']=''
#self.proxy = urllib2.ProxyHandler()
#self.opener = urllib2.urlopen #build_opener(self.proxy)
#urllib2.install_opener(self.opener)
[docs] def test(self):
'''Test downloading the pUC19 plasmid sequence from genbank
returns True if successful. Can be used to test proxy
and other network settings.
'''
try:
result = self.nucleotide("L09137") # pUC19
except urllib2.URLError:
return False
return seguid(result.seq) == "71B4PwSgBZ3htFjJXwHPxtUIPYE"
[docs] def nucleotide(self, item, start=None, stop=None, strand="watson" ):
'''Download a genbank nuclotide record.
Item is a string containing one genbank acession number [#]_
for a nucleotide file. Start and stop are intervals to be
downloaded. This is useful as some genbank records are large.
If strand is "c", "C", "crick", "Crick", "antisense","Antisense",
"2" or 2, the watson(antisense) strand is returned, otherwise
the sense strand is returned.
Alternatively, item can be a string containing an url that returns a
sequence in genbank or FASTA format.
The genbank E-utilities can provide such urls [#]_.
Result is returned as a Dseqrecord object.
Genbank nucleotide accession numbers have this format:
| A12345 = 1 letter + 5 numerals
| AB123456 = 2 letters + 6 numerals
References
----------
.. [#] http://www.dsimb.inserm.fr/~fuchs/M2BI/AnalSeq/Annexes/Sequences/Accession_Numbers.htm
.. [#] http://www.ncbi.nlm.nih.gov/books/NBK25499/#chapter4.EFetch
'''
if item.lower().startswith(("ftp","http")):
try:
url = urlparse(item)
except (AttributeError, TypeError):
url = urlparse("")
if url.scheme:
return read( urllib2.urlopen(item).read() )
matches =((1, re.search("(REGION:\s(?P<start>\d+)\.\.(?P<stop>\d+))", item)),
(2, re.search("(REGION: complement\((?P<start>\d+)\.\.(?P<stop>\d+)\))",item)),
(1, re.search(":(?P<start>\d+)-(?P<stop>\d+)",item)),
(2, re.search(":c(?P<start>\d+)-(?P<stop>\d+)",item)),
(0, None))
# NM_005546 REGION: 1..100
# NM_005546 REGION: complement(1..100)
# 21614549:1-100
# 21614549:c100-1
# http://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?db=nuccore&id=21614549&strand=1&seq_start=1&seq_stop=100&rettype=fasta&retmode=text
# http://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?db=nuccore&id=21614549&strand=2&seq_start=1&seq_stop=100&rettype=fasta&retmode=text
# http://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?db=nuccore&id=21614549&strand=1&seq_start=1&seq_stop=100&rettype=gb&retmode=text
# http://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?db=nuccore&id=21614549&strand=2&seq_start=1&seq_stop=100&rettype=gb&retmode=text
for strand, match in matches:
if match:
start = match.group("start")
stop = match.group("stop")
item = item[:match.start()]
break
# if not strand:
# raise Exception("\nACESSION string is malformed!\n"
# "NM_005546 REGION: 1..100\n"
# "NM_005546 REGION: complement(1..100)\n"
# "21614549:1-100\n"
# "21614549:c100-1\n")
if str(strand).lower() in ("c","crick", "antisense", "2"):
strand = 2
else:
strand = 1
Entrez.email = self.email
return read(Entrez.efetch(db ="nucleotide",
id = item,
rettype = "gb",
seq_start = start,
seq_stop = stop,
strand = strand,
retmode = "text").read())
if __name__=="__main__":
import doctest
doctest.testmod()
gb = Genbank("bjornjobbb@gmail.com")
print gb.email
gene = gb.nucleotide("X59465")