biopy_seq3.py

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Tue Nov 28 12:56:51 2017

@author: tp53
'''

'''
http://biopython.org/DIST/docs/api/Bio.SeqUtils-module.html

seq3(seq, custom_map=None, undef_code='Xaa')
	source code 

Turn a one letter code protein sequence into one with three letter codes.

The single required input argument 'seq' should be a protein sequence using single letter codes, either as a python string or as a Seq or MutableSeq object.

This function returns the amino acid sequence as a string using the three letter amino acid codes. Output follows the IUPAC standard (including ambiguous characters B for "Asx", J for "Xle" and X for "Xaa", and also U for "Sel" and O for "Pyl") plus "Ter" for a terminator given as an asterisk. Any unknown character (including possible gap characters), is changed into 'Xaa' by default.

e.g.

>>> from Bio.SeqUtils import seq3
>>> seq3("MAIVMGRWKGAR*")
'MetAlaIleValMetGlyArgTrpLysGlyAlaArgTer'

You can set a custom translation of the codon termination code using the dictionary "custom_map" argument (which defaults to {'*': 'Ter'}), e.g.

>>> seq3("MAIVMGRWKGAR*", custom_map={"*": "***"})
'MetAlaIleValMetGlyArgTrpLysGlyAlaArg***'

You can also set a custom translation for non-amino acid characters, such as '-', using the "undef_code" argument, e.g.

>>> seq3("MAIVMGRWKGA--R*", undef_code='---')
'MetAlaIleValMetGlyArgTrpLysGlyAla------ArgTer'

If not given, "undef_code" defaults to "Xaa", e.g.

>>> seq3("MAIVMGRWKGA--R*")
'MetAlaIleValMetGlyArgTrpLysGlyAlaXaaXaaArgTer'

This function was inspired by BioPerl's seq3.
"""

from Bio.SeqUtils import seq3
seq3("MAIVMGRWKGAR*")

from Bio.SeqUtils import seq1
seq1("MetAlaIleValMetGlyArgTrpLysGlyAlaArgTer")

seq1("MetAlaIleValMetGlyArgTrpLysGlyAlaArg***", custom_map={"***": "*"})
seq1("MetAlaIleValMetGlyArgTrpLysGlyAla------ArgTer", undef_code='?')

'''
You can set a custom translation of the codon termination code using the dictionary "custom_map" argument (defaulting to {'Ter': '*'}), e.g.

>>> seq1("MetAlaIleValMetGlyArgTrpLysGlyAlaArg***", custom_map={"***": "*"})
'MAIVMGRWKGAR*'

You can also set a custom translation for non-amino acid characters, such as '-', using the "undef_code" argument, e.g.

>>> seq1("MetAlaIleValMetGlyArgTrpLysGlyAla------ArgTer", undef_code='?')
'MAIVMGRWKGA??R*'

If not given, "undef_code" defaults to "X", e.g.

>>> seq1("MetAlaIleValMetGlyArgTrpLysGlyAla------ArgTer")
'MAIVMGRWKGAXXR*'
'''

import rpy2.robjects as robjects

dir(robjects)

from rpy2.robjects.packages import importr
# import R's "base" package
base = importr('base')

# import R's "utils" package
utils = importr('utils')

dir(base)
type(base)

dir(utils)
type(utils)

# select a mirror for R packages
utils.chooseCRANmirror(ind=1) # select the first mirror in the list

# R package names
#packnames = ('ggplot2', 'hexbin')
packnames = ('devtools')

# R vector of strings
from rpy2.robjects.vectors import StrVector

import rpy2.robjects.packages as rpackages

print(rpackages.isinstalled
#print(importr.isinstalled)

# Selectively install what needs to be install.
# We are fancy, just because we can.

#code block below, gives synytax error in the line below,
#maybe try using the 'in' operator
'''
      names_to_install = [x for packnames if not rpackages.isinstalled(x)]
if len(names_to_install) > 0:
    utils.install_packages(StrVector(names_to_install))
'''