def __init__(self, badStrs=None, trueStrs=None, falseStrs=None):
if badStrs is None:
self.badStrs = set()
else:
self.badStrs = set(
[s.lower() for s in SeqUtil.asCollection(badStrs)])
if trueStrs is None:
self.trueStrs = set(
[v.lower() for v in _TrueValues if hasattr(v, "lower")])
self.trueStrs -= self.badStrs
else:
self.trueStrs = set(
[s.lower() for s in SeqUtil.asCollection(trueStrs)])
if self.trueStrs & self.badStrs:
raise ValueError(
"One or more bad values and true values overlap")
if falseStrs is None:
self.falseStrs = set(
[v.lower() for v in _FalseValues if hasattr(v, "lower")])
self.falseStrs -= self.badStrs
else:
self.falseStrs = set(
[s.lower() for s in SeqUtil.asCollection(falseStrs)])
if self.falseStrs & self.badStrs:
raise ValueError(
"One or more bad values and true values overlap")
if self.trueStrs & self.falseStrs:
raise ValueError("One or more true and false values overlap")
def test_findlonglen_returns_longest_length(self):
test_dict = {
'length1': 15,
'length10': 150,
'length100': 1,
}
self.assertEqual(len('length100'), SeqUtil.findlonglen(test_dict))
def __init__(self, badStrs=None, trueStrs=None, falseStrs=None):
if badStrs is None:
self.badStrs = set()
else:
self.badStrs = set([s.lower() for s in SeqUtil.asCollection(badStrs)])
if trueStrs is None:
self.trueStrs = set([v.lower() for v in _TrueValues if hasattr(v, "lower")])
self.trueStrs -= self.badStrs
else:
self.trueStrs = set([s.lower() for s in SeqUtil.asCollection(trueStrs)])
if self.trueStrs & self.badStrs:
raise ValueError("One or more bad values and true values overlap")
if falseStrs is None:
self.falseStrs = set([v.lower() for v in _FalseValues if hasattr(v, "lower")])
self.falseStrs -= self.badStrs
else:
self.falseStrs = set([s.lower() for s in SeqUtil.asCollection(falseStrs)])
if self.falseStrs & self.badStrs:
raise ValueError("One or more bad values and true values overlap")
if self.trueStrs & self.falseStrs:
raise ValueError("One or more true and false values overlap")
inputFile = './data/5.leaderboard_data-1.txt'
#inputFile = 'C:/Users/Ashis/Downloads/dataset_102_4 (1).txt'
outputFile = './results/5.leaderboard.txt'
aminoAcidMassMapFile = './data/integer_mass_table.txt'
# set current directory
os.chdir(curDir)
# read input
with open(inputFile) as f:
N = int(f.readline().strip())
spectrumLine = f.readline().strip()
spectrum = [int(n) for n in spectrumLine.split(" ")]
# load SeqUtil methods
sys.path.append('code')
import SeqUtil
importlib.reload(SeqUtil)
### original work ####
# load amion acid mass
AAMassMap = SeqUtil.loadAminoAcidMass(aminoAcidMassMapFile)
# generate sequence from spectrum
seq = SeqUtil.leaderboardCyclopeptideSequencing(spectrum, N,
AAMassMap.values())
# output
with open(outputFile, "w") as f:
f.writelines("-".join([str(s) for s in seq]))
inputFile = './data/3.theoretical-spectrum-data-1.txt'
#inputFile = 'C:/Users/Ashis/Downloads/dataset_98_3.txt'
outputFile = './results/3.theoretical-spectrum.txt'
aminoAcidMassMapFile = './data/integer_mass_table.txt'
# set current directory
os.chdir(curDir)
# read input
with open(inputFile) as f:
peptide = f.readline().strip()
# load SeqUtil methods
sys.path.append('code')
import SeqUtil
importlib.reload(SeqUtil)
### original work ####
# load amion acid mass
AAMassMap = SeqUtil.loadAminoAcidMass(aminoAcidMassMapFile)
# generate spectrum
spectrum = SeqUtil.cyclospectrum(peptide, AAMassMap)
# output
with open(outputFile, "w") as f:
f.writelines(" ".join([str(s) for s in spectrum]))
def __init__(self, badStrs=("NaN", "?")):
if not SeqUtil.isCollection(badStrs):
self.badStrs = set([badStrs.lower()])
else:
self.badStrs = set([bs.lower() for bs in badStrs])
import os, sys
import SeqUtil, Report
if not os.path.exists('aligns'):
os.mkdir('aligns')
if not os.path.exists('Bayes'):
os.mkdir('Bayes')
#if not os.path.exists('ML'):
# os.mkdir('ML')
out= sys.argv[1]
query = sys.argv[2]
SeqUtil.rename('Data/bac-'+out+'.fas')
if not os.path.exists('aligns/bac-'+out+'.best.nex'):
os.system('prank -d=Data/bac-'+out+' -o=aligns/bac-'+out+' -f=nexus -quiet')
SeqUtil.bayesinNex('aligns/bac-'+out+'.best.nex')
#SeqUtil.splicealign('aligns/bac-'+out+'.best.nex','Bayes/bac-'+out+'-mod.nxs')
#models=SeqUtil.bestmod('Bayes/bac-'+out+'-mod.nxs')
models_ori=SeqUtil.bestmod('aligns/bac-'+out+'.best.nex')
if not os.path.exists('Bayes/bac-'+out+'-bayes.nxs'):
SeqUtil.bayesfile('aligns/bac-'+out+'.best.nex',models_ori,'Bayes/bac-'+out+'-bayes.nxs')
#SeqUtil.bayesfile('Bayes/bac-'+out+'-mod.nxs',models,'Bayes/bac-'+out+'-bayes.nxs')
os.system('mb Bayes/bac-'+out+'-bayes.nxs')
#SeqUtil.pamlseqnex('Bayes/bac-'+out+'-mod.nxs','ML/bac-'+out)
#for mod in models.keys():
# SeqUtil.pamlinput('ML/bac-'+out,'ML/bac-'+out+'.out','ML/bac-'+out+'.ctl',{models.keys()[mod].split('+')[0]:models[models.keys()[mod]][1]})
# os.system('codeml ML/bac-'+out+'.ctl')
# SeqUtil.extractMLtree('ML/bac-'+out+'.out')
Report.generateReport(out,query,models_ori,'bac')
def __init__(self, badStrs=("NaN", "?")):
if not SeqUtil.isCollection(badStrs):
self.badStrs = set([badStrs.lower()])
else:
self.badStrs = set([bs.lower() for bs in badStrs])
# settings
curDir = 'E:/Copy/Coursera/Bioinformatics Algorithms (part-I)/MyPrograms/week2'
#curDir = 'D:/Copy/Coursera/Bioinformatics Algorithms (part-I)/MyPrograms/week2'
inputFile = './data/3.theoretical-spectrum-data-1.txt'
#inputFile = 'C:/Users/Ashis/Downloads/dataset_98_3.txt'
outputFile = './results/3.theoretical-spectrum.txt'
aminoAcidMassMapFile = './data/integer_mass_table.txt'
# set current directory
os.chdir(curDir)
# read input
with open(inputFile) as f:
peptide = f.readline().strip()
# load SeqUtil methods
sys.path.append('code')
import SeqUtil
importlib.reload(SeqUtil)
### original work ####
# load amion acid mass
AAMassMap = SeqUtil.loadAminoAcidMass(aminoAcidMassMapFile)
# generate spectrum
spectrum = SeqUtil.cyclospectrum(peptide, AAMassMap)
# output
with open(outputFile, "w") as f:
f.writelines(" ".join([str(s) for s in spectrum]))
aminoAcidMassMapFile = './data/integer_mass_table.txt'
# set current directory
os.chdir(curDir)
# read input
with open(inputFile) as f:
spectrumLine = f.readline().strip()
spectrum = [int(n) for n in spectrumLine.split(" ")]
# load SeqUtil methods
sys.path.append('code')
import SeqUtil
importlib.reload(SeqUtil)
### original work ####
# load amion acid mass
AAMassMap = SeqUtil.loadAminoAcidMass(aminoAcidMassMapFile)
# generate sequence from spectrum
conv = SeqUtil.spectralConvolution(spectrum)
conv = sorted(conv)
# output
with open(outputFile, "w") as f:
f.writelines(" ".join([str(m) for m in conv]))
aminoAcidMassMapFile = './data/integer_mass_table.txt'
# set current directory
os.chdir(curDir)
# read input
with open(inputFile) as f:
N = int(f.readline().strip())
spectrumLine = f.readline().strip()
spectrum = [int(n) for n in spectrumLine.split(" ")]
# load SeqUtil methods
sys.path.append('code')
import SeqUtil
importlib.reload(SeqUtil)
### original work ####
# load amion acid mass
AAMassMap = SeqUtil.loadAminoAcidMass(aminoAcidMassMapFile)
# generate sequence from spectrum
seq = SeqUtil.leaderboardCyclopeptideSequencing(spectrum, N, AAMassMap.values())
# output
with open(outputFile, "w") as f:
f.writelines("-".join([str(s) for s in seq]))
# set current directory
os.chdir(curDir)
# read input
with open(inputFile) as f:
spectrumLine = f.readline().strip()
spectrum = [int(n) for n in spectrumLine.split(" ")]
# load SeqUtil methods
sys.path.append('code')
import SeqUtil
importlib.reload(SeqUtil)
### original work ####
# load amion acid mass
AAMassMap = SeqUtil.loadAminoAcidMass(aminoAcidMassMapFile)
# generate sequence from spectrum
sequences = SeqUtil.cyclopeptideSequencing(spectrum, AAMassMap.values())
# output
with open(outputFile, "w") as f:
seqStrings = []
for seq in sequences:
seqStrings.append("-".join(str(mass) for mass in seq))
f.writelines(" ".join([s for s in seqStrings]))
inputFile = './data/7.convolutional_seq_data-1.txt'
#inputFile = 'C:/Users/Ashis/Downloads/dataset_104_7.txt'
outputFile = './results/7.convolutional_seq.txt'
# set current directory
os.chdir(curDir)
# read input
with open(inputFile) as f:
M = int(f.readline().strip())
N = int(f.readline().strip())
spectrumLine = f.readline().strip()
spectrum = [int(n) for n in spectrumLine.split(" ")]
# load SeqUtil methods
sys.path.append('code')
import SeqUtil
importlib.reload(SeqUtil)
### original work ####
# generate sequence from spectrum
seq = SeqUtil.convolutionCyclopeptideSequencing(spectrum, M, N)
# output
with open(outputFile, "w") as f:
f.writelines("-".join([str(s) for s in seq]))
# load SeqUtil methods
sys.path.append('code')
import SeqUtil
importlib.reload(SeqUtil)
lenTargetPeptide = len(target)
lenTargetDna = lenTargetPeptide*3
foundDnas = [] # storage of target dan sequences
## search amino acids in forward direction
for start in range(0,3): # loop for reading frame start
curDna = dna[start:]
curRna = SeqUtil.dna2rna(curDna)
peptides = SeqUtil.rna2peptide(curRna, mapFile='./data/RNA_codon_table_1.txt')
pepStartIndex = 0
for pep in peptides:
# find positions of target in peptide
targetPositions = [pos for pos in range(0, len(pep)) if pep[pos:(pos+lenTargetPeptide)]==target]
# get corresponding dna sequence and save
targetDnas = [curDna[(pepStartIndex+pos*3):(pepStartIndex+pos*3+lenTargetDna)] for pos in targetPositions]
foundDnas.extend(targetDnas)
# update pepStartIndex for the next peptide
pepStartIndex = pepStartIndex + (len(pep)+1)*3
if not os.path.exists('aligns'):
os.mkdir('aligns')
if not os.path.exists('Bayes'):
os.mkdir('Bayes')
#if not os.path.exists('ML'):
# os.mkdir('ML')
out=sys.argv[1]
query=sys.argv[2]
try:
paml=sys.argv[3]
paml= paml=='-y'
except IndexError:
paml=False
print "Beginning alignment"
SeqUtil.rename('Data/all-'+out+'.fas')
os.system('prank -d=Data/all-'+out+' -o=aligns/all-'+out+' -f=nexus -quiet')
SeqUtil.bayesinNex('aligns/all-'+out+'.best.nex')
#SeqUtil.splicealign('aligns/all-'+out+'.best.nex','Bayes/all-'+out+'-mod.nxs')
print "Alignment complete.\nCalculating best model for tree finding"
models_ori=SeqUtil.bestmod('aligns/all-'+out+'.best.nex')
#models=SeqUtil.bestmod('Bayes/all-'+out+'-mod.nxs')
#print models_ori, models
if paml:
for mod in models.keys():
SeqUtil.pamlseqnex('Bayes/all-'+out+'-mod.nxs','ML/all-'+out+mod.split('+')[0])
if models[mod][0]=='0' and models[mod][1]=='0':
os.system('phyml -i ML/all-'+out+mod.split('+')[0]+' -d aa -b 100 -m '+mod.split('+')[0]+
' -f e -s BEST -u aligns/all-'+out+'.ed.2.dnd -o tl')
elif models[mod][0]=='0':
os.system('phyml -i '+'ML/all-'+out+mod.split('+')[0]+' -d aa -b 100 -m '+mod.split('+')[0]+
