def parser(variant_string, variant):
# for amino acids only the 3 char is acceptable which is the preferred method
# according to HGVS. eg: For Tryptophan only 'Tyr' is acceptable, not 'Y'
# since we have already got the v_type, set to position 2 after the period
i = 2
# --- Position/ ---
# ignore the '[' or '('
if variant_string[i] == '[' or variant_string[i] == '(':
i += 1
allele = True
# if '?' unknown or '=' no change expected
if variant_string[i] == '?' or variant_string[i] == '=':
variant.position = variant_string[i]
i += 1
variant.operator_value = variant_string[i:]
else:
# get the position amino acid
if (i+3) < len(variant_string):
# get the first amino acid
if variant_string[i] == '*':
first_amino_acid = '*'
else:
first_amino_acid = variant_string[i:i+3]
if not first_amino_acid.isalpha():
#check if first char is alphabet
first_amino_acid = variant_string[i]
i += 1
else:
i += 3
# assume that it is a position value for now
variant.position = first_amino_acid
# get the position of the amino acid, which will be stored
# in the postion intron for now
start_position = i
while i < len(variant_string):
if check_numeric_value(variant_string[i]):
i += 1
else:
break
# get the position value
position = variant_string[start_position:i]
# assume that it is a position intron value for now
variant.position_intron = position
if len(variant_string) > i:
if variant_string[i] == '_':
i += 1
if len(variant_string) > i:
# set the first amino acid to range lower
variant.position = ''
variant.range_lower = first_amino_acid;
# set postion intron value to range lower intron
variant.position_intron = ''
variant.range_lower_intron = position
# get the range upper or second amino acid value
if variant_string[i] == '*':
variant.range_upper = '*'
else:
if variant_string[i:i+3].isalpha():
variant.range_upper = variant_string[i:i+3]
i += 3
else:
varaint.range_upper = variant_string[i]
i += 1
# get the range upper intron
start_position = i
while i < len(variant_string):
if check_numeric_value(variant_string[i]):
i += 1
else:
break
variant.range_upper_intron = variant_string[start_position:i]
# get the operator - the amino acid that has changed
operator = variant_string[i:len(variant_string)]
operator_value = ''
if operator[0] == '*':
operator = '*'
elif 'delins' in operator.lower():
operator = 'delins'
operator_value = variant_string[i+6:len(variant_string)]
elif variant_string[i] == '(':
operator = variant_string[i:len(variant_string)]
operator_value = ''
elif operator.lower() in ('fs*', 'fsx', 'fster'):
f = operator.index('fs*')
operator_value = operator[f:]
operator = operator[0:f]
else:
if variant_string[i:i+3].isalpha():
operator = variant_string[i:i+3]
operator_value = variant_string[i+3:len(variant_string)]
else:
operator = variant_string[i]
operator_value = variant_string[i+1:len(variant_string)]
# set the operator
variant.operator = operator
variant.operator_value = operator_value
return variant
def validate(variant):
# Please not that since protein nomenclature is different to genomic/cDNA structure the values are
# stored differently in the VariantName Class.
# Since the protein positions and ranges contain the amino acid and numeric value for the index
# of that amino acid the variant position, range_lower and range_upper will store the amino acids
# while the intron fields will store the index.
# position: amino acid
if variant.position.strip() != '':
if variant.position not in ('?', '='):
if not variant.position.lower() in ValidValues.amino_acids:
if not variant.position.lower(
) in ValidValues.amino_acids_single:
return False
else:
return True
# position: index of the amino acid
if variant.position_intron.strip() != '':
# if position has a value then intron should too
if variant.position.strip() != '':
if not check_numeric_value(variant.position_intron):
return False
else:
return False
# range_lower: amino acid
if variant.range_lower.strip() != '':
if variant.position.strip() == '':
if not variant.range_lower.lower() in ValidValues.amino_acids:
return False
else:
return False
# range_lower: index of the amino acid
if variant.range_lower_intron.strip() != '':
if variant.range_lower.strip() != '':
if not check_numeric_value(variant.range_lower_intron):
return False
else:
return False
# range_upper: amino acid
if variant.range_upper.strip() != '':
if variant.range_lower.strip() != '':
if not variant.range_upper.lower() in ValidValues.amino_acids:
return False
else:
return False
# range_upper: index of the amino acid
if variant.range_upper_intron.strip() != '':
if variant.range_upper.strip() != '':
if not check_numeric_value(variant.range_upper_intron):
return False
else:
return False
# Operator
if variant.operator.strip() != '':
# check for repeating range
if not variant.operator[0] != '(' or not variant.operator[0] != '[':
if not get_repeater_value(variant.operator):
return False
# check for indels --> 'delins' and insertions --> 'ins'
elif variant.operator.lower() not in ValidValues.protein_operators:
# check for amino acids
if not variant.operator.lower() in ValidValues.amino_acids:
if not variant.operator.lower(
) in ValidValues.amino_acids_single:
return False
else:
return False
# Operator Value: should only contain amino acids for indel and insertion
# operators.
if variant.operator_value.strip() != '':
# frameshifts
if variant.operator_value[0:2] == 'fs':
if len(variant.operator_value) > 2:
if variant.operator_value[2].lower() in ('*', 'x'):
# check for '];['
if '];[' in variant.operator_value[3:]:
p = variant.operator_value.index('];[')
if variant.operator_value[3:p] != '':
if not variant.operator_value[3:p].isdigit():
return False
else:
if not variant.operator_value[3:].isdigit():
return False
else:
return False
else:
# ignore if operator begins with ']' or ')'
if ']' not in variant.operator_value and ')' not in variant.operator_value:
# operator value can not be empty if the operator is an indel or insertion
if variant.operator.lower(
) not in ValidValues.protein_operators:
# The length of operator string should be divisble by 3 since the amino acid
# codes should only be 3 chars long.
if not len(variant.operator_value) % 3 != 0:
return False
else:
# need to check each amino acid if valid
amino_acids = split(variant.operator_value.lower(), 3)
for amino_acid in amino_acids:
item_found = False
if amino_acid in ValidValues.amino_acids:
item_found = True
# if amino acid not found
if not item_found:
return False
return True
def parser(variant_string, variant):
# for amino acids only the 3 char is acceptable which is the preferred method
# according to HGVS. eg: For Tryptophan only 'Tyr' is acceptable, not 'Y'
# since we have already got the v_type, set to position 2 after the period
i = 2
# --- Position/ ---
# ignore the '[' or '('
if variant_string[i] == '[' or variant_string[i] == '(':
i += 1
allele = True
# if '?' unknown or '=' no change expected
if variant_string[i] == '?' or variant_string[i] == '=':
variant.position = variant_string[i]
i += 1
variant.operator_value = variant_string[i:]
else:
# get the position amino acid
if (i + 3) < len(variant_string):
# get the first amino acid
if variant_string[i] == '*':
first_amino_acid = '*'
else:
first_amino_acid = variant_string[i:i + 3]
if not first_amino_acid.isalpha():
#check if first char is alphabet
first_amino_acid = variant_string[i]
i += 1
else:
i += 3
# assume that it is a position value for now
variant.position = first_amino_acid
# get the position of the amino acid, which will be stored
# in the postion intron for now
start_position = i
while i < len(variant_string):
if check_numeric_value(variant_string[i]):
i += 1
else:
break
# get the position value
position = variant_string[start_position:i]
# assume that it is a position intron value for now
variant.position_intron = position
if len(variant_string) > i:
if variant_string[i] == '_':
i += 1
if len(variant_string) > i:
# set the first amino acid to range lower
variant.position = ''
variant.range_lower = first_amino_acid
# set postion intron value to range lower intron
variant.position_intron = ''
variant.range_lower_intron = position
# get the range upper or second amino acid value
if variant_string[i] == '*':
variant.range_upper = '*'
else:
if variant_string[i:i + 3].isalpha():
variant.range_upper = variant_string[i:i + 3]
i += 3
else:
varaint.range_upper = variant_string[i]
i += 1
# get the range upper intron
start_position = i
while i < len(variant_string):
if check_numeric_value(variant_string[i]):
i += 1
else:
break
variant.range_upper_intron = variant_string[
start_position:i]
# get the operator - the amino acid that has changed
operator = variant_string[i:len(variant_string)]
operator_value = ''
if operator[0] == '*':
operator = '*'
elif 'delins' in operator.lower():
operator = 'delins'
operator_value = variant_string[i + 6:len(variant_string)]
elif variant_string[i] == '(':
operator = variant_string[i:len(variant_string)]
operator_value = ''
elif operator.lower() in ('fs*', 'fsx', 'fster'):
f = operator.index('fs*')
operator_value = operator[f:]
operator = operator[0:f]
else:
if variant_string[i:i + 3].isalpha():
operator = variant_string[i:i + 3]
operator_value = variant_string[i +
3:len(variant_string)]
else:
operator = variant_string[i]
operator_value = variant_string[i +
1:len(variant_string)]
# set the operator
variant.operator = operator
variant.operator_value = operator_value
return variant
def validate(variant):
# Please not that since protein nomenclature is different to genomic/cDNA structure the values are
# stored differently in the VariantName Class.
# Since the protein positions and ranges contain the amino acid and numeric value for the index
# of that amino acid the variant position, range_lower and range_upper will store the amino acids
# while the intron fields will store the index.
# position: amino acid
if variant.position.strip() != '':
if variant.position not in ('?', '='):
if not variant.position.lower() in ValidValues.amino_acids:
if not variant.position.lower() in ValidValues.amino_acids_single:
return False
else:
return True
# position: index of the amino acid
if variant.position_intron.strip() != '':
# if position has a value then intron should too
if variant.position.strip() != '':
if not check_numeric_value(variant.position_intron):
return False
else:
return False
# range_lower: amino acid
if variant.range_lower.strip() != '':
if variant.position.strip() == '':
if not variant.range_lower.lower() in ValidValues.amino_acids:
return False
else:
return False
# range_lower: index of the amino acid
if variant.range_lower_intron.strip() != '':
if variant.range_lower.strip() != '':
if not check_numeric_value(variant.range_lower_intron):
return False
else:
return False
# range_upper: amino acid
if variant.range_upper.strip() != '':
if variant.range_lower.strip() != '':
if not variant.range_upper.lower() in ValidValues.amino_acids:
return False
else:
return False
# range_upper: index of the amino acid
if variant.range_upper_intron.strip() != '':
if variant.range_upper.strip() != '':
if not check_numeric_value(variant.range_upper_intron):
return False
else:
return False
# Operator
if variant.operator.strip() != '':
# check for repeating range
if not variant.operator[0] != '(' or not variant.operator[0] != '[':
if not get_repeater_value(variant.operator):
return False
# check for indels --> 'delins' and insertions --> 'ins'
elif variant.operator.lower() not in ValidValues.protein_operators:
# check for amino acids
if not variant.operator.lower() in ValidValues.amino_acids:
if not variant.operator.lower() in ValidValues.amino_acids_single:
return False
else:
return False
# Operator Value: should only contain amino acids for indel and insertion
# operators.
if variant.operator_value.strip() != '':
# frameshifts
if variant.operator_value[0:2] == 'fs':
if len(variant.operator_value) > 2:
if variant.operator_value[2].lower() in ('*', 'x'):
# check for '];['
if '];[' in variant.operator_value[3:]:
p = variant.operator_value.index('];[')
if variant.operator_value[3:p] != '':
if not variant.operator_value[3:p].isdigit():
return False
else:
if not variant.operator_value[3:].isdigit():
return False
else:
return False
else:
# ignore if operator begins with ']' or ')'
if ']' not in variant.operator_value and ')' not in variant.operator_value:
# operator value can not be empty if the operator is an indel or insertion
if variant.operator.lower() not in ValidValues.protein_operators:
# The length of operator string should be divisble by 3 since the amino acid
# codes should only be 3 chars long.
if not len(variant.operator_value) % 3 != 0:
return False
else:
# need to check each amino acid if valid
amino_acids = split(variant.operator_value.lower(), 3)
for amino_acid in amino_acids:
item_found = False
if amino_acid in ValidValues.amino_acids:
item_found = True
# if amino acid not found
if not item_found:
return False
return True
