def ReturnBlossum62Dict():
from Bio.SubsMat.MatrixInfo import blosum62
subMat = {}
for p, v in blosum62.items():
subMat[p] = v
subMat[p[::-1]] = v
return (subMat)
def blosum(msa, ref):
"""
blosum score: sum of BLOSUM(ref, L) for L in all ligands.
Normalized by the number of sums and min/max blos62 score
:param msa:
:param ref:
:return:
"""
Mb, mb = max([v for k, v in blosum62.items()
]), min([v for k, v in blosum62.items()])
out = []
reflen = len(msa.loc[ref])
for i in range(reflen):
norm = 0
res_list = list(msa.iloc[:, i])
reference_res = str(msa.iloc[:, i].loc[ref])
if reference_res not in "-.":
# now min and max is residue specific
this_score = -blosum62[
(reference_res, reference_res
)] # starts in debt to cover self reference hit
for res in res_list:
if res not in "-.":
norm += 1
# exception to catch stupid triangular matrix
try:
bloscore = blosum62[(reference_res, res)]
except KeyError:
bloscore = blosum62[(res, reference_res)]
this_score += bloscore
# print(", ".join([reference_res, "\t"] + res_list + [str(round(this_freq,2))]))
out.append(np.round(((this_score / norm) - mb) / (Mb - mb), 2))
return out
dest='dMutRate',
help='Enter the pct of AA\'s to mutate. Examples: .03, .10, ,... ')
args = parser.parse_args()
# ---------------------------------------------------------------
# constants
# ---------------------------------------------------------------
c_strGoodAAs = re.sub("[BOJUXZ]", "", uppercase)
c_dictSelfScore = {}
c_dictSelfScore["X"] = 0
c_dictdictExchangeScores = {}
for (strAA1, strAA2), fValue in blosum62.items():
if strAA1 in c_strGoodAAs and strAA2 in c_strGoodAAs:
if strAA1 != strAA2:
c_dictdictExchangeScores.setdefault(strAA1,
{})[strAA2] = math.exp(fValue)
c_dictdictExchangeScores.setdefault(strAA2,
{})[strAA1] = math.exp(fValue)
else:
# higher self score should be mutated less, hence negative
c_dictSelfScore[strAA1] = math.exp(-fValue)
# convert exchanges to weighted chooser objects
c_dictChoosers = {}
for strAA, dictWeights in c_dictdictExchangeScores.items():
c_dictChoosers[strAA] = weighted_chooser(dictWeights)
# process codon table
import matplotlib
matplotlib.use("Agg")
from matplotlib import pyplot as pl
import numpy as np
import maps
import standards
DGSolCommand = "dgsol"
gap = "_"
# Amino acid alphabet
aa = "ACDEFGHIKLMNPQRSTVWY" + gap
b62.update({(b, a): v for (a, b), v in b62.items()}) # make BLOSUM62 symmetric
def rank(array):
order = array.argsort()
ranks = np.empty(len(array), dtype=int)
ranks[order] = np.arange(len(array))
return ranks
def triangle_inequality(matrix):
# NOTE: only works on square matrices
if isinstance(matrix, dict):
N = set()
for (k1, k2) in matrix.keys():
N.add(k1)
from Bio.Seq import Seq
from Bio.SubsMat.MatrixInfo import blosum62
from Bio.Data import CodonTable
from zopy.stats import WeightedChooser
from zopy.dictation import col2dict
# ---------------------------------------------------------------
# constants
# ---------------------------------------------------------------
c_good_aas = re.sub("[BOJUXZ]", "", uppercase)
c_self_score = {}
c_exchange_scores = {}
for (aa1, aa2), value in blosum62.items():
if aa1 in c_good_aas and aa2 in c_good_aas:
if aa1 != aa2:
c_exchange_scores.setdefault(aa1, {})[aa2] = math.exp(value)
c_exchange_scores.setdefault(aa2, {})[aa1] = math.exp(value)
else:
# higher self score should be mutated less, hence negative
c_self_score[aa1] = math.exp(-value)
# convert exchanges to WeightedChooser objects
c_choosers = {}
for aa, weights in c_exchange_scores.items():
c_choosers[aa] = WeightedChooser(weights)
# process codon table
c_codons = {}
for codon, aa in CodonTable.unambiguous_dna_by_name[
#! /usr/bin/env python3.2
from homPolDist import writeBothToFile, fastaParser, stripLowerCase
from subprocess import call, check_output
from Bio.Seq import translate
from math import log10
from numpy import median,std
from Bio.SubsMat.MatrixInfo import blosum62 as blosum
import binascii
import sys
blosum62 = {}
for (a,b),val in blosum.items():
oldT = (a,b)
newT = (b,a)
blosum62[tuple(newT)] = val
blosum62[tuple(oldT)] = val
randGeneFN = "randGene.fasta"
transGeneFN = "transGene.fasta"
# Trans gene no homopolymer errors
tnhpReadsFN = "tnhpReads.fasta"
randSffFN = "randFlow.sff"
transSffFN = "transFlow.sff"
randReadsFN = "randReads.fasta"
transReadsFN = "transReads.fasta"
randCodonAlnInFile = "randCodonAlnInFile.fasta"
transCodonAlnInFile = "transCodonAlnInFile.fasta"
tnhpCodonAlnInFile = "tnhpCodonAlnInFile.fasta"
randCodonAlnOutFile = "randCodonAlnOutFile.fasta"
transCodonAlnOutFile = "transCodonAlnOutFile.fasta"
tnhpCodonAlnOutFile = "tnhpCodonAlnOutFile.fasta"