allpy: allpy/dna.py annotate

allpy

annotate allpy/dna.py @ 1168:b556c96c6719

blocks3d/www Makefile: never check certificates of github, they are too confusing for wget

author	Daniil Alexeyevsky <dendik@kodomo.fbb.msu.ru>
date	Mon, 26 May 2014 17:20:29 +0400
parents	afed1fd8920c
children

rev	line source
bnagaev@240	1 import base
me@353	2 import data.codes
dendik@1100	3 from data.genetic_code import standard_dna_code
me@280	4
dendik@382	5 import dna
dendik@1100	6 import protein
dendik@382	7
me@328	8 class Monomer(base.Monomer):
me@328	9 """DNA monomers: nucleotides."""
me@328	10 type = 'dna'
dendik@382	11 types = dna
me@280	12 by_code1 = {}
me@280	13 by_code3 = {}
me@280	14 by_name = {}
me@353	15 Monomer._initialize(data.codes.dna)
bnagaev@240	16
bnagaev@240	17 class Sequence(base.Sequence):
dendik@382	18 types = dna
dendik@382	19
dendik@930	20 def reverse_complemented(self):
dendik@930	21 """Return a new sequence, reverse-complement to self.
dendik@930	22
dendik@930	23 Name of the sequence is name of self with apostrophe added.
dendik@930	24 """
dendik@930	25 complement = {'A': 'T', 'T': 'A', 'C': 'G', 'G': 'C'}
dendik@1091	26 result = self.types.Sequence(
dendik@1091	27 name=self.name +"'",
dendik@1091	28 description=self.description,
dendik@1091	29 source=self.source
dendik@930	30 )
dendik@1091	31 for monomer in reversed(self):
dendik@1091	32 result.append_monomer(complement.get(monomer.code1, 'N'))
dendik@1091	33 return result
dendik@930	34
dendik@1100	35 def translated(self, code=None, name=None, description=None, source=None):
dendik@1100	36 """Return a new protein sequence translated from self.
dendik@1100	37
dendik@1100	38 `code` is a dict of triplet of dna `code1`s -> aminoacid `name`.
dendik@1100	39
dendik@1100	40 If `code` is not specified, the standard genetic code is used.
dendik@1100	41
dendik@1100	42 If `code` is specified, it may contain only the changed codons.
dendik@1100	43
dendik@1100	44 Class of proteins to use is `self.types.protein.Sequence`, you
dendik@1100	45 are free to replace it at will.
dendik@1100	46
dendik@1100	47 Return new protein sequence where:
dendik@1100	48
dendik@1100	49 * `name` is self.name with "_tr" appended
dendik@1100	50 * `description` is self.description with " translated" appended
dendik@1100	51 * `source` is the same as self.source
dendik@1100	52 """
dendik@1100	53 if code:
dendik@1100	54 code, modification = dict(standard_dna_code), code
dendik@1100	55 code.update(modification)
dendik@1100	56 else:
dendik@1100	57 code = standard_dna_code
dendik@1100	58 result = self.types.protein.Sequence([],
dendik@1100	59 name=name or self.name + "_tr",
dendik@1100	60 description=description or self.description + " translated",
dendik@1100	61 source=source or self.source
dendik@1100	62 )
dendik@1100	63 seen_stop = False
dendik@1100	64 for a, b, c in zip(self[::3], self[1::3], self[2::3]):
dendik@1100	65 assert not seen_stop, "Stop-codon must be the last one"
dendik@1100	66 triplet = a.code1 + b.code1 + c.code1
dendik@1100	67 aa_name = code[triplet]
dendik@1100	68 if aa_name == "Stop":
dendik@1100	69 seen_stop = True
dendik@1100	70 continue
dendik@1100	71 result.append_monomer(name=aa_name)
dendik@1100	72 return result
dendik@1100	73
dendik@931	74 class Column(base.Column):
dendik@931	75 types = dna
dendik@931	76
dendik@382	77 class Alignment(base.Alignment):
dendik@382	78 types = dna
bnagaev@240	79
me@341	80 class Block(Alignment, base.Block):
me@341	81 pass
me@341	82
me@280	83 # vim: set ts=4 sts=4 sw=4 et: