def process_snp(snp, full_tree_string):
if sum(1 for x in snp.column if x) < 7:
return '\t'.join(str(x) for x in (snp.variant_id, '*', '*'))
pruned_tree = snp.get_pruned_tree(full_tree_string)
# define the map from the taxon to the amino acid
taxon_aa_pairs = zip(g_ordered_taxon_names, snp.column)
taxon_to_aa_letter = dict((t, aa) for t, aa in taxon_aa_pairs if aa)
# get the weights of the taxa
taxon_weight_pairs = LeafWeights.get_stone_weights(pruned_tree)
# calculate the standardized physicochemical property table
standardized_property_array = MAPP.get_standardized_property_array(
MAPP.g_property_array)
# calculate the physicochemical property correlation matrix
correlation_matrix = MAPP.get_property_correlation_matrix(
standardized_property_array)
# estimate the amino acid distribution for the column,
# taking into account the tree and a uniform prior.
weights = []
aa_indices = []
for taxon, weight in taxon_weight_pairs:
weights.append(weight)
aa_index = aa_letter_to_aa_index(taxon_to_aa_letter[taxon])
aa_indices.append(aa_index)
aa_distribution = MAPP.estimate_aa_distribution(weights, aa_indices)
# estimate the mean and variance of each physicochemical property
est_pc_means = MAPP.estimate_property_means(
standardized_property_array, aa_distribution)
est_pc_variances = MAPP.estimate_property_variances(
standardized_property_array, aa_distribution)
# calculate the deviation from each property mean
# for each possible amino acid
deviations = MAPP.get_deviations(
est_pc_means, est_pc_variances, standardized_property_array)
# calculate the impact scores
impact_scores = MAPP.get_impact_scores(correlation_matrix, deviations)
# calculate the p-values
p_values = []
for score in impact_scores:
ntaxa = len(taxon_weight_pairs)
p_values.append(MAPP.get_p_value(score, ntaxa))
# show the impact score and p-value for the mutant
letter_to_impact = dict(zip(Codon.g_aa_letters, impact_scores))
letter_to_pvalue = dict(zip(Codon.g_aa_letters, p_values))
impact = letter_to_impact[snp.mutant_aa]
pvalue = letter_to_pvalue[snp.mutant_aa]
return '\t'.join(str(x) for x in (snp.variant_id, impact, pvalue))
def process_snp(snp, full_tree_string):
if sum(1 for x in snp.column if x) < 7:
return '\t'.join(str(x) for x in (snp.variant_id, '*', '*'))
pruned_tree = snp.get_pruned_tree(full_tree_string)
# define the map from the taxon to the amino acid
taxon_aa_pairs = zip(g_ordered_taxon_names, snp.column)
taxon_to_aa_letter = dict((t, aa) for t, aa in taxon_aa_pairs if aa)
# get the weights of the taxa
taxon_weight_pairs = LeafWeights.get_stone_weights(pruned_tree)
# calculate the standardized physicochemical property table
standardized_property_array = MAPP.get_standardized_property_array(
MAPP.g_property_array)
# calculate the physicochemical property correlation matrix
correlation_matrix = MAPP.get_property_correlation_matrix(
standardized_property_array)
# estimate the amino acid distribution for the column,
# taking into account the tree and a uniform prior.
weights = []
aa_indices = []
for taxon, weight in taxon_weight_pairs:
weights.append(weight)
aa_index = aa_letter_to_aa_index(taxon_to_aa_letter[taxon])
aa_indices.append(aa_index)
aa_distribution = MAPP.estimate_aa_distribution(weights, aa_indices)
# estimate the mean and variance of each physicochemical property
est_pc_means = MAPP.estimate_property_means(standardized_property_array,
aa_distribution)
est_pc_variances = MAPP.estimate_property_variances(
standardized_property_array, aa_distribution)
# calculate the deviation from each property mean
# for each possible amino acid
deviations = MAPP.get_deviations(est_pc_means, est_pc_variances,
standardized_property_array)
# calculate the impact scores
impact_scores = MAPP.get_impact_scores(correlation_matrix, deviations)
# calculate the p-values
p_values = []
for score in impact_scores:
ntaxa = len(taxon_weight_pairs)
p_values.append(MAPP.get_p_value(score, ntaxa))
# show the impact score and p-value for the mutant
letter_to_impact = dict(zip(Codon.g_aa_letters, impact_scores))
letter_to_pvalue = dict(zip(Codon.g_aa_letters, p_values))
impact = letter_to_impact[snp.mutant_aa]
pvalue = letter_to_pvalue[snp.mutant_aa]
return '\t'.join(str(x) for x in (snp.variant_id, impact, pvalue))
def get_response_content(fs):
# start writing the html response
out = StringIO()
print >> out, '<html>'
print >> out, '<body>'
# get the tree and the column sent by the user
pruned_tree, taxon_to_aa_letter = get_tree_and_column(fs)
# get the weights of the taxa
taxon_weight_pairs = LeafWeights.get_stone_weights(pruned_tree)
# show the raw physicochemical property table
if fs.show_raw_pc_table:
print >> out, 'raw physicochemical property table:'
print >> out, '<br/>'
col_labels = MAPP.g_property_names
row_labels = Codon.g_aa_letters
table = MAPP.g_property_array
print >> out, HtmlTable.get_labeled_table_string(
col_labels, row_labels, table)
print >> out, '<br/><br/>'
# calculate the standardized physicochemical property table
standardized_property_array = MAPP.get_standardized_property_array(
MAPP.g_property_array)
# show the standardized physicochemical property table
if fs.show_standardized_pc_table:
print >> out, 'standardized physicochemical property table:'
print >> out, '<br/>'
col_labels = MAPP.g_property_names
row_labels = Codon.g_aa_letters
table = standardized_property_array
print >> out, HtmlTable.get_labeled_table_string(
col_labels, row_labels, table)
print >> out, '<br/><br/>'
# calculate the physicochemical property correlation matrix
correlation_matrix = MAPP.get_property_correlation_matrix(
standardized_property_array)
# show the physicochemical property correlation matrix
if fs.show_pc_correlation_matrix:
print >> out, 'physicochemical property correlation matrix:'
print >> out, '<br/>'
col_labels = MAPP.g_property_names
row_labels = MAPP.g_property_names
table = correlation_matrix
print >> out, HtmlTable.get_labeled_table_string(
col_labels, row_labels, table)
print >> out, '<br/><br/>'
# show the pruned tree
if fs.show_tree:
tree_string = NewickIO.get_narrow_newick_string(pruned_tree, 80)
lines = StringIO(tree_string).readlines()
lines = [line.rstrip() for line in lines]
print >> out, 'pruned phylogenetic tree in newick format:'
print >> out, '<pre>'
for line in lines:
print >> out, cgi.escape(line)
print >> out, '</pre>'
print >> out, '<br/>'
# show the weights
if fs.show_weights:
taxa, weights = zip(*taxon_weight_pairs)
table = [weights]
row_labels = ['weight']
col_labels = taxa
print >> out, 'taxon weights:'
print >> out, '<br/>'
print >> out, HtmlTable.get_labeled_table_string(
col_labels, row_labels, table)
print >> out, '<br/><br/>'
# estimate the amino acid distribution for the column,
# taking into account the tree and a uniform prior.
weights = []
aa_indices = []
for taxon, weight in taxon_weight_pairs:
weights.append(weight)
aa_indices.append(aa_letter_to_aa_index(taxon_to_aa_letter[taxon]))
aa_distribution = MAPP.estimate_aa_distribution(weights, aa_indices)
# show the estimated amino acid distribution
if fs.show_aa_distribution:
table = [aa_distribution]
row_labels = ['weight']
col_labels = Codon.g_aa_letters
print >> out, 'estimated amino acid distribution:'
print >> out, '<br/>'
print >> out, HtmlTable.get_labeled_table_string(
col_labels, row_labels, table)
print >> out, '<br/><br/>'
# estimate the mean and variance of each physicochemical property
est_pc_means = MAPP.estimate_property_means(
standardized_property_array, aa_distribution)
est_pc_variances = MAPP.estimate_property_variances(
standardized_property_array, aa_distribution)
# show the estimated mean and variance of each physicochemical property
if fs.show_pc_distribution:
table = [est_pc_means, est_pc_variances]
row_labels = ['mean', 'variance']
col_labels = MAPP.g_property_names
print >> out, 'estimated physicochemical property moments:'
print >> out, '<br/>'
print >> out, HtmlTable.get_labeled_table_string(
col_labels, row_labels, table)
print >> out, '<br/><br/>'
# calculate the deviation from each property mean
# for each possible amino acid
deviations = MAPP.get_deviations(
est_pc_means, est_pc_variances, standardized_property_array)
# show the deviation from each property mean for each possible amino acid
if fs.show_deviations:
print >> out, 'deviations of amino acids from the normal distribution'
print >> out, 'estimated for each property:'
print >> out, '<br/>'
col_labels = MAPP.g_property_names
row_labels = Codon.g_aa_letters
table = deviations
print >> out, HtmlTable.get_labeled_table_string(
col_labels, row_labels, table)
print >> out, '<br/><br/>'
# calculate the impact scores
impact_scores = MAPP.get_impact_scores(correlation_matrix, deviations)
# show the impact scores
if fs.show_impact_scores:
table = [impact_scores]
row_labels = ['impact']
col_labels = Codon.g_aa_letters
print >> out, 'impact scores:'
print >> out, '<br/>'
print >> out, HtmlTable.get_labeled_table_string(
col_labels, row_labels, table)
print >> out, '<br/><br/>'
# calculate the p-values
p_values = []
for score in impact_scores:
ntaxa = len(taxon_weight_pairs)
p_values.append(MAPP.get_p_value(score, ntaxa))
# show the p-values
if fs.show_p_values:
table = [p_values]
row_labels = ['p-value']
col_labels = Codon.g_aa_letters
print >> out, 'p-values:'
print >> out, '<br/>'
print >> out, HtmlTable.get_labeled_table_string(
col_labels, row_labels, table)
print >> out, '<br/><br/>'
# write the html footer
print >> out, '</body>'
print >> out, '</html>'
# return the response
return out.getvalue()
def get_response_content(fs):
# get the upper case amino acid letter
aa_of_interest = fs.aminoacid.upper()
if aa_of_interest not in Codon.g_aa_letters:
raise HandlingError('invalid amino acid: ' + fs.aminoacid)
# get the newick tree
tree = NewickIO.parse(fs.tree, Newick.NewickTree)
# get the alignment
out = StringIO()
try:
finder = KGEA.Finder(g_index_dir, g_valid_chromosome_strings_pathname,
g_fasta_dir)
# note that some of these amino acids can be gaps
taxon_aa_pairs = list(
finder.gen_taxon_aa_pairs(fs.chromosome, fs.position))
lines = [taxon + '\t' + aa for taxon, aa in taxon_aa_pairs]
if taxon_aa_pairs:
# get the map from the taxon to the amino acid
taxon_to_aa_letter = dict((taxon, aa)
for taxon, aa in taxon_aa_pairs
if aa in Codon.g_aa_letters)
selected_taxa = set(taxon_to_aa_letter)
ntaxa = len(selected_taxa)
# assert that we have enough taxa
# to calculate the p-value from the MAPP statistic
mintaxa = 7
if ntaxa < mintaxa:
raise HandlingError(
'this column has only %d aligned amino acids but we want at least %d'
% (ntaxa, mintaxa))
# modify the tree so that we keep only the taxa of interest
modify_tree(tree, selected_taxa)
# get the taxon weights from the tree
taxon_weight_pairs = LeafWeights.get_stone_weights(tree)
# calculate the standardized physicochemical property table
standardized_property_array = MAPP.get_standardized_property_array(
MAPP.g_property_array)
# calculate the physicochemical property correlation matrix
correlation_matrix = MAPP.get_property_correlation_matrix(
standardized_property_array)
# estimate the amino acid distribution for the column
weights = []
aa_indices = []
for taxon, weight in taxon_weight_pairs:
weights.append(weight)
aa_indices.append(
aa_letter_to_aa_index(taxon_to_aa_letter[taxon]))
aa_distribution = MAPP.estimate_aa_distribution(
weights, aa_indices)
# estimate the mean and variance of each physicochemical property
est_pc_means = MAPP.estimate_property_means(
standardized_property_array, aa_distribution)
est_pc_variances = MAPP.estimate_property_variances(
standardized_property_array, aa_distribution)
# calculate the deviation from each property mean for each possible amino acid
deviations = MAPP.get_deviations(est_pc_means, est_pc_variances,
standardized_property_array)
# calculate the impact scores
impact_scores = MAPP.get_impact_scores(correlation_matrix,
deviations)
# calculate the p-values
p_values = [
MAPP.get_p_value(score, ntaxa) for score in impact_scores
]
# show the p-value of the amino acid of interest
print >> out, 'MAPP p-value:'
print >> out, p_values[aa_letter_to_aa_index(aa_of_interest)]
print >> out
# show the MAPP statistic of the amino acid of interest
print >> out, 'MAPP statistic:'
print >> out, impact_scores[aa_letter_to_aa_index(aa_of_interest)]
print >> out
# show the aligned column
print >> out, 'aligned column:'
print >> out, '\n'.join(lines)
else:
print >> out, 'no aligned amino acids were found at this position'
except KGEA.KGEAError as e:
print >> out, e
return out.getvalue()
def get_response_content(fs):
# get the upper case amino acid letter
aa_of_interest = fs.aminoacid.upper()
if aa_of_interest not in Codon.g_aa_letters:
raise HandlingError('invalid amino acid: ' + fs.aminoacid)
# get the newick tree
tree = NewickIO.parse(fs.tree, Newick.NewickTree)
# get the alignment
out = StringIO()
try:
finder = KGEA.Finder(
g_index_dir, g_valid_chromosome_strings_pathname, g_fasta_dir)
# note that some of these amino acids can be gaps
taxon_aa_pairs = list(
finder.gen_taxon_aa_pairs(fs.chromosome, fs.position))
lines = [taxon + '\t' + aa for taxon, aa in taxon_aa_pairs]
if taxon_aa_pairs:
# get the map from the taxon to the amino acid
taxon_to_aa_letter = dict((taxon, aa) for taxon, aa in taxon_aa_pairs if aa in Codon.g_aa_letters)
selected_taxa = set(taxon_to_aa_letter)
ntaxa = len(selected_taxa)
# assert that we have enough taxa
# to calculate the p-value from the MAPP statistic
mintaxa = 7
if ntaxa < mintaxa:
raise HandlingError('this column has only %d aligned amino acids but we want at least %d' % (ntaxa, mintaxa))
# modify the tree so that we keep only the taxa of interest
modify_tree(tree, selected_taxa)
# get the taxon weights from the tree
taxon_weight_pairs = LeafWeights.get_stone_weights(tree)
# calculate the standardized physicochemical property table
standardized_property_array = MAPP.get_standardized_property_array(MAPP.g_property_array)
# calculate the physicochemical property correlation matrix
correlation_matrix = MAPP.get_property_correlation_matrix(standardized_property_array)
# estimate the amino acid distribution for the column
weights = []
aa_indices = []
for taxon, weight in taxon_weight_pairs:
weights.append(weight)
aa_indices.append(aa_letter_to_aa_index(taxon_to_aa_letter[taxon]))
aa_distribution = MAPP.estimate_aa_distribution(weights, aa_indices)
# estimate the mean and variance of each physicochemical property
est_pc_means = MAPP.estimate_property_means(standardized_property_array, aa_distribution)
est_pc_variances = MAPP.estimate_property_variances(standardized_property_array, aa_distribution)
# calculate the deviation from each property mean for each possible amino acid
deviations = MAPP.get_deviations(est_pc_means, est_pc_variances, standardized_property_array)
# calculate the impact scores
impact_scores = MAPP.get_impact_scores(correlation_matrix, deviations)
# calculate the p-values
p_values = [MAPP.get_p_value(score, ntaxa) for score in impact_scores]
# show the p-value of the amino acid of interest
print >> out, 'MAPP p-value:'
print >> out, p_values[aa_letter_to_aa_index(aa_of_interest)]
print >> out
# show the MAPP statistic of the amino acid of interest
print >> out, 'MAPP statistic:'
print >> out, impact_scores[aa_letter_to_aa_index(aa_of_interest)]
print >> out
# show the aligned column
print >> out, 'aligned column:'
print >> out, '\n'.join(lines)
else:
print >> out, 'no aligned amino acids were found at this position'
except KGEA.KGEAError as e:
print >> out, e
return out.getvalue()
def get_response_content(fs):
# start writing the html response
out = StringIO()
print >> out, '<html>'
print >> out, '<body>'
# get the tree and the column sent by the user
pruned_tree, taxon_to_aa_letter = get_tree_and_column(fs)
# get the weights of the taxa
taxon_weight_pairs = LeafWeights.get_stone_weights(pruned_tree)
# show the raw physicochemical property table
if fs.show_raw_pc_table:
print >> out, 'raw physicochemical property table:'
print >> out, '<br/>'
col_labels = MAPP.g_property_names
row_labels = Codon.g_aa_letters
table = MAPP.g_property_array
print >> out, HtmlTable.get_labeled_table_string(
col_labels, row_labels, table)
print >> out, '<br/><br/>'
# calculate the standardized physicochemical property table
standardized_property_array = MAPP.get_standardized_property_array(
MAPP.g_property_array)
# show the standardized physicochemical property table
if fs.show_standardized_pc_table:
print >> out, 'standardized physicochemical property table:'
print >> out, '<br/>'
col_labels = MAPP.g_property_names
row_labels = Codon.g_aa_letters
table = standardized_property_array
print >> out, HtmlTable.get_labeled_table_string(
col_labels, row_labels, table)
print >> out, '<br/><br/>'
# calculate the physicochemical property correlation matrix
correlation_matrix = MAPP.get_property_correlation_matrix(
standardized_property_array)
# show the physicochemical property correlation matrix
if fs.show_pc_correlation_matrix:
print >> out, 'physicochemical property correlation matrix:'
print >> out, '<br/>'
col_labels = MAPP.g_property_names
row_labels = MAPP.g_property_names
table = correlation_matrix
print >> out, HtmlTable.get_labeled_table_string(
col_labels, row_labels, table)
print >> out, '<br/><br/>'
# show the pruned tree
if fs.show_tree:
tree_string = NewickIO.get_narrow_newick_string(pruned_tree, 80)
lines = StringIO(tree_string).readlines()
lines = [line.rstrip() for line in lines]
print >> out, 'pruned phylogenetic tree in newick format:'
print >> out, '<pre>'
for line in lines:
print >> out, cgi.escape(line)
print >> out, '</pre>'
print >> out, '<br/>'
# show the weights
if fs.show_weights:
taxa, weights = zip(*taxon_weight_pairs)
table = [weights]
row_labels = ['weight']
col_labels = taxa
print >> out, 'taxon weights:'
print >> out, '<br/>'
print >> out, HtmlTable.get_labeled_table_string(
col_labels, row_labels, table)
print >> out, '<br/><br/>'
# estimate the amino acid distribution for the column,
# taking into account the tree and a uniform prior.
weights = []
aa_indices = []
for taxon, weight in taxon_weight_pairs:
weights.append(weight)
aa_indices.append(aa_letter_to_aa_index(taxon_to_aa_letter[taxon]))
aa_distribution = MAPP.estimate_aa_distribution(weights, aa_indices)
# show the estimated amino acid distribution
if fs.show_aa_distribution:
table = [aa_distribution]
row_labels = ['weight']
col_labels = Codon.g_aa_letters
print >> out, 'estimated amino acid distribution:'
print >> out, '<br/>'
print >> out, HtmlTable.get_labeled_table_string(
col_labels, row_labels, table)
print >> out, '<br/><br/>'
# estimate the mean and variance of each physicochemical property
est_pc_means = MAPP.estimate_property_means(standardized_property_array,
aa_distribution)
est_pc_variances = MAPP.estimate_property_variances(
standardized_property_array, aa_distribution)
# show the estimated mean and variance of each physicochemical property
if fs.show_pc_distribution:
table = [est_pc_means, est_pc_variances]
row_labels = ['mean', 'variance']
col_labels = MAPP.g_property_names
print >> out, 'estimated physicochemical property moments:'
print >> out, '<br/>'
print >> out, HtmlTable.get_labeled_table_string(
col_labels, row_labels, table)
print >> out, '<br/><br/>'
# calculate the deviation from each property mean
# for each possible amino acid
deviations = MAPP.get_deviations(est_pc_means, est_pc_variances,
standardized_property_array)
# show the deviation from each property mean for each possible amino acid
if fs.show_deviations:
print >> out, 'deviations of amino acids from the normal distribution'
print >> out, 'estimated for each property:'
print >> out, '<br/>'
col_labels = MAPP.g_property_names
row_labels = Codon.g_aa_letters
table = deviations
print >> out, HtmlTable.get_labeled_table_string(
col_labels, row_labels, table)
print >> out, '<br/><br/>'
# calculate the impact scores
impact_scores = MAPP.get_impact_scores(correlation_matrix, deviations)
# show the impact scores
if fs.show_impact_scores:
table = [impact_scores]
row_labels = ['impact']
col_labels = Codon.g_aa_letters
print >> out, 'impact scores:'
print >> out, '<br/>'
print >> out, HtmlTable.get_labeled_table_string(
col_labels, row_labels, table)
print >> out, '<br/><br/>'
# calculate the p-values
p_values = []
for score in impact_scores:
ntaxa = len(taxon_weight_pairs)
p_values.append(MAPP.get_p_value(score, ntaxa))
# show the p-values
if fs.show_p_values:
table = [p_values]
row_labels = ['p-value']
col_labels = Codon.g_aa_letters
print >> out, 'p-values:'
print >> out, '<br/>'
print >> out, HtmlTable.get_labeled_table_string(
col_labels, row_labels, table)
print >> out, '<br/><br/>'
# write the html footer
print >> out, '</body>'
print >> out, '</html>'
# return the response
return out.getvalue()
