def chi_goodness_yats(self):
observed = np.asarray(self.observed)
expected = np.asarray(self.expected)
observed = observed + 0.5 * np.sign(expected - observed)
chi2_yats, p_yats = power_divergence(observed, expected, ddof=0)
add_result(self, "chi2_yats", round(chi2_yats, 5))
add_result(self, "p_yats", round(p_yats, 5))
def transcribed_sequence(self):
self.get_seq_names_and_contents()
for seq_name, seq in zip(self.seq_names, self.seq_contents):
raw_seq = Seq(seq, IUPAC.unambiguous_dna)
transformed_seq = raw_seq.transcribe()
add_result(self, seq_name, str(transformed_seq))
def reverse_sequence(self):
self.get_seq_names_and_contents()
for seq_name, seq in zip(self.seq_names, self.seq_contents):
raw_seq = Seq(seq, IUPAC.unambiguous_dna)
transformed_seq = str(raw_seq[::-1])
add_result(self, seq_name, transformed_seq)
def reverse_and_complement_sequences(self):
self.get_seq_names_and_contents()
for seq_name, seq in zip(self.seq_names, self.seq_contents):
raw_seq = Seq(seq, IUPAC.unambiguous_dna)
transformed_seq = raw_seq.reverse_complement()
add_result(self, seq_name, str(transformed_seq))
def calculate(self):
# check if type is float change math formula
if isinstance(self.data["amplified_marker"], float):
fi = self.data["amplified_marker"]
else:
fi = int(self.data["amplified_marker"]) / (
int(self.data["amplified_marker"]) +
int(self.data["absence_marker"]))
pic = 1 - (fi**2 + ((1 - fi)**2))
add_result(self, "PIC", round(pic, 4))
return self.results
def correlation(self):
field_sum = float(self.data["field_sum"])
chi2_standard, p_standard, dof, ex = self.chi_square_standard()
chi2_yats, p_yats = self.chi_square_yats()
if dof == 1:
correlation_standard = sqrt(chi2_standard / field_sum)
correlation_yats = sqrt(chi2_yats / field_sum)
add_result(self, "coefficient of contingency type", "Phi")
else:
m = min(self.data["width"], self.data["height"])
correlation_standard = sqrt(chi2_standard / (field_sum * (m - 1)))
correlation_yats = sqrt(chi2_yats / (field_sum * (m - 1)))
add_result(self, "coefficient of contingency type", "Crammer`s V")
return chi2_standard, p_standard, dof, chi2_yats, p_yats, correlation_standard, correlation_yats
def raw_sequence(self):
filename = create_seq_file(self.data['sequences'])
threshold = self.data.get('threshold', 0.55)
muscle = MuscleCommandline(input=filename)
stdout, stderr = muscle()
align = AlignIO.read(StringIO(stdout), "fasta")
summary_align = AlignInfo.SummaryInfo(align)
consensus = summary_align.gap_consensus(threshold=threshold,
ambiguous='N')
add_result(self, "Consensus sequence", str(consensus))
add_result(self, "Sequence length", len(consensus))
remove_temp_file(filename)
return self.results
def genebank_seq(self):
Entrez.email = "*****@*****.**"
with Entrez.efetch(db="nucleotide",
rettype="gb",
retmode="text",
id=self.data['seq-name-1']) as handle:
seq_record = SeqIO.read(handle, "gb")
seq_name_1 = str(seq_record.id) + " " + str(
seq_record.description)[:25]
self.data['seq-content-1'] = str(seq_record.seq)
self.data['seq-name-1'] = seq_name_1
with Entrez.efetch(db="nucleotide",
rettype="gb",
retmode="text",
id=self.data['seq-name-2']) as handle:
seq_record = SeqIO.read(handle, "gb")
seq_name_2 = str(seq_record.id) + " " + str(
seq_record.description)[:25]
self.data['seq-content-2'] = str(seq_record.seq)
self.data['seq-name-2'] = seq_name_2
alignment = self.get_alignments()
add_result(self, f'Seq id. {seq_name_1} ... length [bp]',
len(self.data['seq-content-1']))
add_result(self, f'Seq id. {seq_name_2} ... length [bp]',
len(self.data['seq-content-2']))
ident = alignment.count("|")
coverage = self.get_coverage()
average_identity = self.get_average_identity(ident)
frag_identity = self.get_frag_identity(ident)
add_result(self, "Coverage [%]", coverage)
add_result(self, "Average identity [%]", average_identity)
add_result(self, "Fragmental identity [%]", frag_identity)
return self.results
def raw_sequence(self):
self.remove_newlines()
seq_name_1 = self.data['seq-name-1']
seq_name_2 = self.data['seq-name-2']
seq_content_1 = self.data['seq-content-1']
seq_content_2 = self.data['seq-content-2']
alignment = self.get_alignments()
add_result(self, f'Sequence {seq_name_1} length', len(seq_content_1))
add_result(self, f'Sequence {seq_name_2} length', len(seq_content_2))
ident = alignment.count("|")
coverage = self.get_coverage()
average_identity = self.get_average_identity(ident)
frag_identity = self.get_frag_identity(ident)
add_result(self, "Coverage [%]", coverage)
add_result(self, "Average identity [%]", average_identity)
add_result(self, "Fragmental identity [%]", frag_identity)
return self.results
def calculate(self):
alfa = self.data["alfa"]
ho, he, rho, e_ho, e_he, e_rho, p, q = self.calculate_expected_observed(
)
chi, pval = chisquare([ho, he, rho], f_exp=[e_ho, e_he, e_rho])
add_result(self, 'expected number of homozygotes', round(e_ho, 2))
add_result(self, 'expected number of heterozygotes', round(e_he, 2))
add_result(self, 'expected number of rare homozygotes',
round(e_rho, 2))
add_result(self, 'p', round(p, 5))
add_result(self, 'q', round(q, 5))
add_result(self, 'p-value', round(pval, 5))
add_result(self, 'Chi-square value', round(chi, 5))
if ho < 5 or he < 5 or rho < 5:
chi_yates, pval_yates = self.calculate_yates_correction()
add_result(self, 'Yate`s chi-square value', round(chi_yates, 5))
add_result(self, 'Yate`s p-value', round(pval_yates, 5))
# set pval to pval_yates for next if statement ~JANO
pval = pval_yates
if pval <= alfa:
msg = (
"Distribution does not consistent with Hardy Weinberg's law at the level "
"of significance: {}".format(alfa))
add_result(self, 'status', msg)
fis = 1 - (he / e_he)
add_result(self, 'fis', round(fis, 4))
elif pval > alfa:
msg = "Distribution consistent with Hardy Weinberg's law at the level of significance: {}".format(
alfa)
add_result(self, 'status', msg)
return self.results
def calculate(self):
add_result(self, "H", self.calculate_h())
add_result(self, "PIC", self.calculate_pic())
return self.results
def calculate(self):
chi2_standard, p_standard, dof, chi2_yats, p_yats, correlation_standard, correlation_yats = self.correlation(
)
add_result(self, "dof", dof)
add_result(self, "chi2_standard", round(chi2_standard, 5))
add_result(self, "p_standard", round(p_standard, 5))
add_result(self, "correlation_standard", round(correlation_standard,
5))
add_result(self, "chi2_yats", round(chi2_yats, 5))
add_result(self, "p_yats", round(p_yats, 5))
add_result(self, "correlation_yats", round(correlation_yats, 5))
return self.results
def chi_goodness_standard(self):
chi2, p = chisquare(self.observed, f_exp=self.expected)
add_result(self, "chi2_standard", round(chi2, 5))
add_result(self, "p_standard", round(p, 5))
add_result(self, "dof", len(self.observed) - 1)
