def calculate_physiochemical_features(temp_dict, sequence):
analyzed_seq = ProteinAnalysis(sequence)
charge_at_pH7 = analyzed_seq.charge_at_pH(7)
instability_index = analyzed_seq.instability_index()
molecular_weight = analyzed_seq.molecular_weight()
aromaticity = analyzed_seq.aromaticity()
molar_extinction_coefficient = analyzed_seq.molar_extinction_coefficient()
range_l, range_h = molar_extinction_coefficient
molar_extinction_coefficient = (float(range_l) + float(range_h)) / 2
gravy = analyzed_seq.gravy(
) #Grand Average Hyrdopathy - Higher value = More Hydrophobic
isoelectric_point = analyzed_seq.isoelectric_point()
helix_fraction, turn_fraction, sheet_fraction = analyzed_seq.secondary_structure_fraction(
)
physiochem_dict = {
"Charge at pH7": charge_at_pH7,
"Instability Index": instability_index,
"Molecular Wt": molecular_weight,
"Aromaticity": aromaticity,
"Molar Extinction Coeff": molar_extinction_coefficient,
"Gravy": gravy,
"Isoelectric pt": isoelectric_point,
"Helix Fraction": helix_fraction,
"Turn Fraction": turn_fraction,
"Sheet Fraction": sheet_fraction
}
temp_dict.update(physiochem_dict)
#Adding separately because get_amino_acids_percent() generates a dictionary on its own
aa_percent = analyzed_seq.get_amino_acids_percent()
temp_dict.update(aa_percent)
def physchem_props(data):
"""Calculate the physicochemical properties per protein in ara_d."""
new_table = []
header = "ID\tclass\tindex\tsequon\tsequence\tmol_weight\tgravy\taromaticity\tinstab_index\tiso_point\n"
new_table.append(header)
for line in data:
split_line = line.rstrip().split('\t')
seq = split_line[-2] # Sequon, not sequence
# Calculates the properties
if "X" in seq or '*' in seq or seq == '':
continue # Skip non-usable sequences, only negs
try:
a_seq = ProteinAnalysis(seq)
# Update ara_d with new physchem properties
results = [
a_seq.molecular_weight(),
a_seq.gravy(),
a_seq.aromaticity(),
a_seq.instability_index(),
#a_seq.flexibility(),
a_seq.isoelectric_point(),
#a_seq.secondary_structure_fraction(),
]
except:
print(split_line)
sys.exit(1)
new_line = line.rstrip() + "\t{}\t{}\t{}\t{}\t{}\n".format(*results)
new_table.append(new_line)
return new_table
def sequence_vector(temp_window: str, window: int = 6, chemical=1):
"""
This vector takes the sequence and has each amino acid represented by an int
0 represents nonstandard amino acids or as fluff for tails/heads of sequences
Strip is a list which can be modified as user needs call for
"""
temp_window = clean(temp_window)
temp_window = windower(sequence=temp_window, position=int(len(temp_window)*.5), wing_size=window)
vec = []
aa = {"G": 1, "A": 2, "L": 3, "M": 4, "F": 5, "W": 6, "K": 7, "Q": 8, "E": 9, "S": 10, "P": 11, "V": 12, "I": 13,
"C": 14, "Y": 15, "H": 16, "R": 17, "N": 18, "D": 19, "T": 20, "X": 0}
for i in temp_window:
vec.append(aa[i])
if len(vec) != (window*2)+1:
t = len(vec)
for i in range((window*2)+1-t):
vec.append(0)
# Hydrophobicity is optional
if chemical == 1:
s = ProteinAnalysis(temp_window)
vec.append(s.gravy())
vec.append(s.instability_index())
vec.append(s.aromaticity())
return vec
def protein_analysis():
if session.username == None:
redirect(URL(r=request, c='account', f='log_in'))
from Bio.SeqUtils.ProtParam import ProteinAnalysis
form = FORM(
TABLE(
TR(
"Amino acid sequence: ",
TEXTAREA(_type="text",
_name="sequence",
requires=IS_NOT_EMPTY())),
INPUT(_type="submit", _value="SUBMIT")))
if form.accepts(request.vars, session):
session['sequence'] = seqClean(form.vars.sequence.upper())
X = ProteinAnalysis(session['sequence'])
session['aa_count'] = X.count_amino_acids()
session['percent_aa'] = X.get_amino_acids_percent()
session['mw'] = X.molecular_weight()
session['aromaticity'] = X.aromaticity()
session['instability'] = X.instability_index()
session['flexibility'] = X.flexibility()
session['pI'] = X.isoelectric_point()
session['sec_struct'] = X.secondary_structure_fraction()
redirect(URL(r=request, f='protein_analysis_output'))
return dict(form=form)
def protParam(seq):
params = ProteinAnalysis(seq)
mw = params.molecular_weight()
c_aa = params.count_amino_acids()
p_aa = params.get_amino_acids_percent()
gravy = params.gravy()
aromaticity = params.aromaticity()
isoelectric_point = params.isoelectric_point()
ext_coeff = sum([c_aa["W"]*5690,c_aa["Y"]*1280,c_aa["C"]*120])
mgml = ext_coeff * (1./mw)
print("Amino acid count")
pprint.pprint(c_aa)
print("Amino acid percent")
pprint.pprint(p_aa)
print("Molecular weight")
print("%f Da"%mw)
print("Gravy")
print(gravy)
print("Isoelectric point")
print(isoelectric_point)
print("Aromaticity")
print(aromaticity)
print("Extinction coefficient: %d M-1cm-1 (Assuming reduced)"%ext_coeff)
print("")
def parse_pro_sequence(self, p_seq, id=None, desc=None):
try:
p_seq = ''.join([pro for pro in p_seq if pro in proteins])
# append fasta sequence metadata
self.id.append(id)
self.description.append(desc)
# reverse translate protein to nucleotide sequence
n_seq = ''.join([list(dna_codons.keys())[list(dna_codons.values()).index(pro)] for pro in p_seq])
self.nucleotide_sequence.append(n_seq)
self.protein_sequence.append(p_seq)
# self.protein_sequence.append(str(record.seq.translate()).replace('*', ' '))
# GC content
self.gc_content.append(self.calculate_gc_content(n_seq))
# protein analysis methods
analysis = ProteinAnalysis(p_seq)
self.amino_acid_dict.append(analysis.get_amino_acids_percent())
self.molecular_weight.append(analysis.molecular_weight())
self.instability_index.append(analysis.instability_index())
self.aromaticity.append(analysis.aromaticity())
except Exception as e:
print('-'*80)
print(f"Exception in parsing uploaded virus sequence: {e}")
traceback.print_exc(file=sys.stdout)
print('-'*80)
def feat_extract(sequences):
list_dict_feat = []
for sequence in sequences:
protein = ProteinAnalysis(sequence)
sequence_feat = defaultdict(float)
sequence_len = len(sequence)
sequence_feat["sequence_length"] = sequence_len
sequence_feat["aromaticty"] = protein.aromaticity()
sequence_feat["isoeletric_point"] = protein.isoelectric_point()
#sequence_feat["flexibility"] = protein.flexibility()
if ('X' not in sequence) and ('O' not in sequence) and (
'U' not in sequence) and ('B' not in sequence):
sequence_feat["molecular_weight"] = protein.molecular_weight()
for letter in sequence:
sequence_feat["relative_fre_{}".format(letter)] += 1 / sequence_len
for property in dic_properties:
if letter in dic_properties[property]:
sequence_feat['freq_{}'.format(property)] += 1
for letter in sequence[0:50]:
sequence_feat["relative_fre_start{}".format(letter)] += 1 / 50
for letter in sequence[-51:-1]:
sequence_feat["relative_fre_end{}".format(letter)] += 1 / 50
list_dict_feat.append(sequence_feat)
return list_dict_feat
def my_own_filtering(input_file, output_file, filt_gc=45, filt_arom=0.01):
sequences = {}
c = 0
with open(input_file, "r") as content:
for record in SeqIO.parse(content, "fasta"):
c += 1
# calculate GC content using Bio
calc_gc = SeqUtils.GC(record.seq)
# calculate aromaticity using Bio
prot_seq = record.seq.translate()
X = ProteinAnalysis(str(prot_seq))
calc_arom = X.aromaticity()
# so, now you can filter
if calc_gc >= filt_gc and calc_arom >= filt_arom:
sequences[record.id] = record.se
# write a new fasta file with aminoacids
records = []
for seq_id, seq in sequences.items():
records.append(SeqRecord(seq.translate(), id=seq_id, description=""))
write_file = open('my_fasta', 'w')
SeqIO.write(records, write_file, 'fasta')
write_file.close()
# print the percentage
print(len(records) / c)
def protAnalysis(self, content):
result, resultFlexDic = dict(), dict()
content = Parsers.normalizeSequence(content, self.sourceType)
protein = ProteinAnalysis(content)
result['proteinMWeight'] = protein.molecular_weight()
result['proteinAroma'] = protein.aromaticity()
result['proteinInstab'] = protein.instability_index()
result['proteinIsoelec'] = protein.isoelectric_point()
result['proteinGravy'] = protein.gravy()
proteinStructure = protein.secondary_structure_fraction()
protStruct = self.flatten('proteinSecstruc', proteinStructure)
result = {**protStruct, **result}
# merge result and protein Structure
flexibility = protein.flexibility()
flexibFlat = self.flatten('proteinFlex', flexibility)
flexibAmino = self.flatten(list(content), flexibility)
flattened = {**flexibFlat, **result}
flattenedFlexDic = {**flexibAmino, **result}
return result, flattened, flattenedFlexDic,
def protein_properties(seq):
"""Return a tuple with some protein biochemical properties
seq is a Bio.Seq.Seq or str representing protein sequence
"""
pa = ProteinAnalysis(seq)
aa_counts = pa.count_amino_acids()
arom = pa.aromaticity()
isoelec = pa.isoelectric_point()
try:
instability = pa.instability_index()
except KeyError:
instability = None
try:
gravy = pa.gravy()
except KeyError:
gravy = None
return ProtProp(aa=str(seq),
gravy=gravy,
aromaticity=arom,
isoelectric_point=isoelec,
instability=instability,
aa_counts=aa_counts)
def protParam(seq):
params = ProteinAnalysis(seq)
mw = params.molecular_weight()
c_aa = params.count_amino_acids()
p_aa = params.get_amino_acids_percent()
gravy = params.gravy()
aromaticity = params.aromaticity()
isoelectric_point = params.isoelectric_point()
ext_coeff = sum([c_aa["W"] * 5690, c_aa["Y"] * 1280, c_aa["C"] * 120])
mgml = ext_coeff * (1. / mw)
print("Amino acid count")
pprint.pprint(c_aa)
print("Amino acid percent")
pprint.pprint(p_aa)
print("Molecular weight")
print("%f Da" % mw)
print("Gravy")
print(gravy)
print("Isoelectric point")
print(isoelectric_point)
print("Aromaticity")
print(aromaticity)
print("Extinction coefficient: %d M-1cm-1 (Assuming reduced)" % ext_coeff)
print("")
def featureExtraction(train_df, test_df):
#feature extraction using bio library to acquire peptide attributes
n = len(train_df)
Y = train_df[0]
train_df = train_df.drop(columns=0)
train_df = train_df.rename(columns={1: 0})
big = pd.concat([train_df, test_df], ignore_index=True)
big['molecular_weight'] = 0.0
#big['flexibility'] = 0
big['isoelectric_point'] = 0.0
big['aromaticity'] = 0.0
big['stability'] = 0.0
for i in range(len(big)):
#print(big.iloc[i, 0])
val = big.iloc[i, 0]
#invalid peptide check, set all values to 0
if 'X' in val or 'Z' in val:
big.at[i, 'molecular_weight'] = -1
#big.at[i, 'flexibility'] = -1
big.at[i, 'isoelectric_point'] = -1
big.at[i, 'aromaticity'] = -1
big.at[i, 'stability'] = -1
continue
model = ProteinAnalysis(val)
big.at[i, 'molecular_weight'] = model.molecular_weight()
#big.at[i, 'flexibility'] = model.flexibility()
big.at[i, 'isoelectric_point'] = model.isoelectric_point()
big.at[i, 'aromaticity'] = model.aromaticity()
big.at[i, 'stability'] = model.instability_index()
big = big.drop(columns=0)
train_df = big.iloc[:n, ]
test_df = big.iloc[n:, ]
return train_df, test_df, Y
def _protein_parameters(self, sequence):
"""Calculates physicochemical properties for the amino acid sequence.
Args:
sequence: str, amino acid sequence.
Returns:
property_arr: np array, vector of properties.
"""
analysis = ProteinAnalysis(sequence)
property_arr = []
property_arr.append(analysis.molecular_weight())
property_arr.append(analysis.aromaticity())
property_arr.append(analysis.instability_index())
property_arr.append(analysis.gravy())
property_arr.append(analysis.isoelectric_point())
secondary = analysis.secondary_structure_fraction()
property_arr.append(secondary[0])
property_arr.append(secondary[1])
property_arr.append(secondary[2])
molar_extinction_coefficient = analysis.molar_extinction_coefficient()
property_arr.append(molar_extinction_coefficient[0])
property_arr.append(molar_extinction_coefficient[1])
property_arr.append(self._net_charge(sequence))
return np.array(property_arr)
def prot_param_features(seq):
features = {}
pa = ProteinAnalysis(str(seq.seq)) # .replace('X','G').replace('B','A')
# 1. Amino Acid Percent
aa = pa.get_amino_acids_percent()
aa_dict = {"frac_{}".format(k): v for k, v in aa.items()}
features.update(aa_dict)
# 2. Aromaticity
features["aromaticity"] = pa.aromaticity()
# 3. Isoelectric Point
features["isoelectric"] = pa.isoelectric_point()
# 4. Molecular Weight
try:
features["mol_weight"] = pa.molecular_weight()
except ValueError:
replaced = str(seq.seq).replace('X', 'G').replace('B', 'N')
# 5. Flexibility
# try:
# features["flexibility"] = np.mean(pa.flexibility())
# except KeyError:
# replaced = str(seq.seq).replace('X', 'G').replace('B', 'N').replace('U','C')
# features["flexibility"] = np.mean(ProteinAnalysis(replaced).flexibility())
# 6. Secondary Structure Fraction
struc = ["struc_helix", "struc_turn", "struc_sheet"]
ss = pa.secondary_structure_fraction()
features.update(dict(zip(struc, ss)))
return features
def get_protein_analysis(aa):
protein_analysis = ProteinAnalysis(aa)
analyze = [protein_analysis.molecular_weight(),
protein_analysis.aromaticity(),
protein_analysis.instability_index(),
protein_analysis.isoelectric_point(),
protein_analysis.gravy()] + list(
protein_analysis.secondary_structure_fraction())
return analyze
def get_aromacity(self):
"""
Calculates Aromacity from sequence (1 value) from biopython
:return: dictionary with the value of aromacity
"""
res = {}
analysed_seq = ProteinAnalysis(self.ProteinSequence)
res['Aromacity'] = analysed_seq.aromaticity()
return res
def processSeq(seq):
''' Protein features found:
- Sequence Length
- Amino Acid Composition (global)
- Amino Acid Composition (First 50/Last 50)
- Isoelectric Point
- Aromacity
- Grand Average Hydropathy (Gravy)
- Molecular Weight (global)
- Molecular Weight (First 50/Last 50)
- Secondary Structure Fraction
'''
# seq = str(seq_record.seq)
prot = ProteinAnalysis(seq)
# desc = str(seq_record.description).split('_')
# species = desc[1].split(' ')[0]
seq_length = len(seq)
isoelectric = prot.isoelectric_point()
gravy = calculateGravy(seq, 0, seq_length)
aroma = prot.aromaticity()
ss_frac = prot.secondary_structure_fraction()
mol_global_weight = calculateMolecularWeight(seq, 0, seq_length)
AA_global_dist = getAAPercent(seq, 0, seq_length)
flex_global = calculateFlexibility(seq, 0, seq_length)
if (seq_length > 50):
AA_local_head = getAAPercent(seq, 0, 50)
AA_local_tail = getAAPercent(seq, seq_length - 50, seq_length)
mol_local_weight_head = calculateMolecularWeight(seq, 0, 50)
mol_local_weight_tail = calculateMolecularWeight(
seq, seq_length - 50, seq_length)
flex_localh = calculateFlexibility(seq, 0, 50)
flex_localt = calculateFlexibility(seq, seq_length - 50, seq_length)
else:
AA_local_head = AA_global_dist
AA_local_tail = AA_global_dist
mol_local_weight_head = mol_global_weight
mol_local_weight_tail = mol_global_weight
flex_localh = flex_global
flex_localt = flex_global
return_vector = [seq_length,aroma,
isoelectric,
mol_global_weight,
mol_local_weight_head,
mol_local_weight_tail,
gravy,flex_global,
flex_localh,
flex_localt] + \
AA_global_dist + AA_local_head + AA_local_tail + list(ss_frac)
# print seq_length, GC_distribution, mol_weight, aroma, isoelectric
return return_vector
def find_composition(df_original):
df_copy = df_original.copy()
column_names = []
for ch in codes:
column_names.append(ch + '_percent')
column_names.append(ch + '_percent_first')
column_names.append(ch + '_percent_last')
column_names.append('len')
column_names.append('weight')
column_names.append('gravy')
column_names.append('flex_mean')
column_names.append('flex_std')
column_names.append('ss_helix')
column_names.append('ss_turn')
column_names.append('ss_sheet')
column_names.append('iep')
column_names.append('aromaticity')
df = pd.DataFrame(columns=column_names)
for _, seq in enumerate(tqdm(df_copy['seq'])):
df_temp = pd.Series()
sequence = str(seq)
analysed = ProteinAnalysis(sequence)
analysed_first = ProteinAnalysis(sequence[:first_n])
analysed_last = ProteinAnalysis(sequence[-last_n:])
df_temp['len'] = analysed.length
df_temp['ss_helix'], df_temp['ss_turn'], df_temp['ss_sheet'] = analysed.secondary_structure_fraction()
df_temp['iep'] = analysed.isoelectric_point()
# overall
for aa, percent in analysed.get_amino_acids_percent().items():
df_temp[aa + '_percent'] = percent
# # first N
for aa, percent in analysed_first.get_amino_acids_percent().items():
df_temp[aa + '_percent_first'] = percent
# last N
for aa, percent in analysed_last.get_amino_acids_percent().items():
df_temp[aa + '_percent_last'] = percent
df_temp['weight'] = analysed.molecular_weight()
df_temp['gravy'] = analysed.gravy()
df_temp['aromaticity'] = analysed.aromaticity()
df_temp['flex_mean'] = np.mean(analysed.flexibility())
df_temp['flex_std'] = np.std(analysed.flexibility())
df = df.append(df_temp, ignore_index=True)
return pd.concat([df_copy, df], axis=1)
def biopython_protein_analysis(inseq):
"""Utiize Biopython's ProteinAnalysis module to return general sequence properties of an amino acid string.
For full definitions see: http://biopython.org/DIST/docs/api/Bio.SeqUtils.ProtParam.ProteinAnalysis-class.html
Args:
inseq: Amino acid sequence
Returns:
dict: Dictionary of sequence properties. Some definitions include:
instability_index: Any value above 40 means the protein is unstable (has a short half life).
secondary_structure_fraction: Percentage of protein in helix, turn or sheet
TODO:
Finish definitions of dictionary
"""
inseq = ssbio.protein.sequence.utils.cast_to_str(inseq)
analysed_seq = ProteinAnalysis(inseq)
info_dict = {}
info_dict['amino_acids_content-biop'] = analysed_seq.count_amino_acids()
info_dict[
'amino_acids_percent-biop'] = analysed_seq.get_amino_acids_percent()
info_dict['length-biop'] = analysed_seq.length
info_dict['monoisotopic-biop'] = analysed_seq.monoisotopic
info_dict['molecular_weight-biop'] = analysed_seq.molecular_weight()
info_dict['aromaticity-biop'] = analysed_seq.aromaticity()
info_dict['instability_index-biop'] = analysed_seq.instability_index()
# TODO: What is flexibility?
info_dict['flexibility-biop'] = analysed_seq.flexibility()
info_dict['isoelectric_point-biop'] = analysed_seq.isoelectric_point()
# grand average of hydrophobicity
info_dict['gravy-biop'] = analysed_seq.gravy()
# Separated secondary_structure_fraction into each definition
# info_dict['secondary_structure_fraction-biop'] = analysed_seq.secondary_structure_fraction()
info_dict[
'percent_helix_naive-biop'] = analysed_seq.secondary_structure_fraction(
)[0]
info_dict[
'percent_turn_naive-biop'] = analysed_seq.secondary_structure_fraction(
)[1]
info_dict[
'percent_strand_naive-biop'] = analysed_seq.secondary_structure_fraction(
)[2]
return info_dict
def phyChemProps(seq):
svv = [0 for x in range(10)]
X = ProteinAnalysis(seq)
svv[0] = X.aromaticity()
svv[1] = X.secondary_structure_fraction()[0]
svv[2] = X.secondary_structure_fraction()[1]
svv[3] = X.secondary_structure_fraction()[2]
svv[4] = X.gravy()
svv[5] = X.instability_index()
svv[6] = X.isoelectric_point()
svv[7] = X.molecular_weight()
svv[8] = X.molar_extinction_coefficient()[0]
svv[9] = X.molar_extinction_coefficient()[1]
return svv
def biochemical_properties(sequence: str) -> Dict[str, Any]:
# Define objects used for calculations
analysis_object = ProteinAnalysis(sequence)
descriptor_object = PyPro.GetProDes(sequence)
sequence_object = Seq(sequence)
# TODO(Ahmed): Verify that all these calculations are actually returning reasonable values
# For example, it says the percent composition of every amino acid is zero when I run
# calculate_biochem_properties.biochemical_properties('qwertyipasdfghklcvnm')
return {
'Isoelectric point': analysis_object.isoelectric_point(),
'Molecular weight':
analysis_object.molecular_weight(), # Daltons? Amu? g/mol?
'Aromaticity': analysis_object.aromaticity(),
'Instability index': analysis_object.instability_index(),
'GRAVY': analysis_object.gravy(),
'H-bonding percent': h_bonding_percent(sequence),
'Melting temp': melting_temp(sequence),
'LCC': lcc.lcc_simp(sequence)
}
def calculate_properties_from_sequence(self):
"""
Function to calculate some molecular properties based on RDKit functionalities
Arguments:
Sequence - amino acid sequence of the peptide
Return:
Average Eisenberg hydrophobicity
ProtParam parameters: Isolectric point, aromaticity, instability index, amino acid percentage
"""
# Hydrophobicity -> Eisenberg scale
hydrophobicity = {
'A': 0.620,
'R': -2.530,
'N': -0.780,
'D': -0.900,
'C': 0.290,
'Q': -0.850,
'E': -0.740,
'G': 0.480,
'H': -0.400,
'Y': 0.260,
'I': 1.380,
'L': 1.060,
'K': -1.500,
'M': 0.640,
'F': 1.190,
'P': 0.120,
'S': -0.180,
'T': -0.050,
'W': 0.810,
'V': 1.080
}
self.avg_hydro = sum([hydrophobicity[resi] for resi in self.sequence])
# ProParam properties
prot_parameters = ProteinAnalysis(self.sequence)
self.aromaticity = prot_parameters.aromaticity()
self.aa_percent = prot_parameters.get_amino_acids_percent()
self.instability_index = prot_parameters.instability_index()
self.isoelectric_point = prot_parameters.isoelectric_point()
def __init__(self, sequence):
self.sequence = sequence
self.sequence_length = len(sequence)
analysis = ProteinAnalysis(sequence)
self.amino_acid_percents = analysis.get_amino_acids_percent()
self.amino_acids_composition = calculate_amino_acids_composition(sequence)
self.aromaticity = analysis.aromaticity()
self.instability = analysis.instability_index()
self.flexibility = calculate_flexibility(sequence)
protein_scale_parameters = [{'name': 'Hydrophilicity', 'dictionary': hw},
{'name': 'Surface accessibility', 'dictionary': em},
{'name': 'Janin Interior to surface transfer energy scale', 'dictionary': ja},
{'name': 'Bulkiness', 'dictionary': bulkiness},
{'name': 'Polarity', 'dictionary': polarity},
{'name': 'Buried residues', 'dictionary': buried_residues},
{'name': 'Average area buried', 'dictionary': average_area_buried},
{'name': 'Retention time', 'dictionary': retention_time}]
self.protein_scales = calculate_protein_scales(analysis, protein_scale_parameters)
self.isoelectric_point = analysis.isoelectric_point()
self.secondary_structure_fraction = calculate_secondary_structure_fraction(analysis)
self.molecular_weight = analysis.molecular_weight()
self.kyte_plot = analysis.gravy()
self.pefing = calculate_pefing(sequence)
# next parameters are calculated using R.Peptides
r('require(Peptides)')
r('sequence = "{0}"'.format(sequence))
self.aliphatic_index = r('aindex(sequence)')[0]
self.boman_index = r('boman(sequence)')[0]
self.charges = calculate_charges(sequence, 1.0, 14.0, 0.5, 'Lehninger')
self.hydrophobicity = r('seq(sequence)')[0]
angles = [{'name': 'Alpha-helix', 'angle': -47},
{'name': '3-10-helix', 'angle': -26},
{'name': 'Pi-helix', 'angle': -80},
{'name': 'Omega', 'angle': 180},
{'name': 'Antiparallel beta-sheet', 'angle': 135},
{'name': 'Parallel beta-sheet', 'angle': 113}]
if self.amino_acid_percents['P'] + self.amino_acid_percents['G'] > 0.3:
angles.append({'name': 'Polygly-polypro helix', 'angle': 153})
self.hydrophobic_moments = calculate_hydrophobic_moments(sequence, angles)
self.kidera_factors = calculate_kidera_factors(sequence)
self.peptide_types = calculate_peptide_types(sequence, angles)
def protein_analysis():
if session.username == None: redirect(URL(r=request,f='../account/log_in'))
from Bio.SeqUtils.ProtParam import ProteinAnalysis
form = FORM(TABLE(
TR("Amino acid sequence: ",
TEXTAREA(_type="text", _name="sequence",
requires=IS_NOT_EMPTY())),
INPUT(_type="submit", _value="SUBMIT")))
if form.accepts(request.vars,session):
session['sequence'] = seqClean(form.vars.sequence.upper())
X = ProteinAnalysis(session['sequence'])
session['aa_count'] = X.count_amino_acids()
session['percent_aa'] = X.get_amino_acids_percent()
session['mw'] = X.molecular_weight()
session['aromaticity'] = X.aromaticity()
session['instability'] = X.instability_index()
session['flexibility'] = X.flexibility()
session['pI'] = X.isoelectric_point()
session['sec_struct'] = X.secondary_structure_fraction()
redirect(URL(r=request, f='protein_analysis_output'))
return dict(form=form)
def GRAvy_ARomo(seq, genetic_code_=1, G=False, A=False):
"""calculating Gravy and Aroma for DNA sequence.
Args:
seq (str):DNA sequence
genetic_code_(int): default = 1, The Genetic Codes number described by NCBI (https://www.ncbi.nlm.nih.gov/Taxonomy/Utils/wprintgc.cgi)
G (bool): default = False
A (bool): default = False
Returns:
- Gravy value if arg(G) is True
- Aroma value if arg(A) is True
- None if both args are False
"""
from Bio.SeqUtils.ProtParam import ProteinAnalysis
from Bio.Seq import Seq
try:
seq = Seq(seq)
except:
pass
translate_seq = str(seq.translate(table=genetic_code_))
protein_seq = translate_seq.replace("*", "")
protein_seq = ProteinAnalysis(protein_seq)
AROMO = protein_seq.aromaticity()
gravy = protein_seq.gravy()
if G and G == True:
return gravy
elif A and A == True:
return AROMO
def get_features(seq):
"""get global features from a protein sequence
Parameters
----------
seq : str
protein sequence
Return
----------
dictionary:
global features of the protein sequence
"""
features = {}
features['undefined_count'] = len([x for x in seq if x in ['X','B','Z',"'",'O','U']])
features['length'] = len(seq)
features['perc_undefined_count'] = features['undefined_count']/features['length']
features['entropy'] = entropy(seq)
features['ideal_entropy'] = entropy_ideal(len(seq))
features['perc_entropy'] = features['entropy']/features['ideal_entropy']
features['hydr_count'] = sum(1 for x in seq if x in hydrophobic_proteins)
features['polar_count'] = sum(1 for x in seq if x in polar_proteins)
features['buried'] = sum(buried[x] for x in seq if x in hydrophobic_proteins)
seq = ''.join([x for x in seq if x not in ['X','B','Z',"'",'O','U']])
protein = ProteinAnalysis(seq)
features['gravy'] = protein.gravy()
features['molecular_weight'] = protein.molecular_weight()
features['aromaticity'] = protein.aromaticity()
features['instability_index'] = protein.instability_index()
features['isoelectric_point'] = protein.isoelectric_point()
features['helix'], features['turn'], features['sheet'] = protein.secondary_structure_fraction()
features.update(protein.count_amino_acids())
# features.update(protein.get_amino_acids_percent())
return features
def GetFeatures (My_seq):
Features = {}
ProteinAnalysis(My_seq)
analysed_seq = ProteinAnalysis(My_seq)
#Caracteristicas monovaloradas
Features["Molecular_weight"] = analysed_seq.molecular_weight()
Features["Aromaticity"] = analysed_seq.aromaticity()
Features["Instability_index"] = analysed_seq.instability_index()
Features["Isoelectric_point"] = analysed_seq.isoelectric_point()
#Caracteristicas multivaloradas
Features["Flexibility"] = analysed_seq.flexibility() # List 580
Features["Second_structure_fraction"] = analysed_seq.secondary_structure_fraction() #3 Tupla
Features["Count_amino_acids"] = analysed_seq.count_amino_acids() #20 Dict
Features["Amino_acids_percent"] = analysed_seq.get_amino_acids_percent() #20 Dict
return Features
def seqs_to_features(self, seqs, no_seqs):
""" Extract the features from the sequences."""
X = np.zeros((no_seqs, 32))
for i, s in enumerate(chain(*seqs)): # iterate over all sequences
# get amino acid counts
alphabet = 'ABCDEFGHIKLMNPQRSTUVWXY' # no JOZ
for j, letter in enumerate(alphabet):
X[i, j] = s.count(letter) / len(s)
# other analysis
analysis = ProteinAnalysis(
s.replace('X', 'A').replace('B', 'A').replace('U', 'A'))
X[i, -1] = analysis.molecular_weight()
X[i, -2] = analysis.aromaticity()
X[i, -3] = analysis.instability_index()
X[i, -4] = analysis.isoelectric_point()
helix_array_sheet_fracs = analysis.secondary_structure_fraction()
X[i, -5] = helix_array_sheet_fracs[0]
X[i, -6] = helix_array_sheet_fracs[1]
X[i, -7] = helix_array_sheet_fracs[2]
X[i, -8] = len(s)
X[i, -9] = analysis.gravy() # mean hydrophobicity
return X
def physchem_props(ara_d):
"""Calculate the physicochemical properties per protein in ara_d."""
c = 0
g = 0
for protein in ara_d:
seq = ara_d[protein]["sequence"]
# Calculates the properties
if "X" in seq:
continue # Skip non-usable sequences, only negs
if '*' in seq:
if ara_d[protein]["pos"] != []:
print(protein)
continue
a_seq = ProteinAnalysis(seq)
# Update ara_d with new physchem properties
results = [
a_seq.molecular_weight(),
a_seq.gravy(),
a_seq.aromaticity(),
a_seq.instability_index(),
a_seq.flexibility(),
a_seq.isoelectric_point(),
a_seq.secondary_structure_fraction(),
]
keys = [
"mol_weight",
"gravy",
"aromaticity",
"instab_index",
"flexi",
"iso_point",
"seq_struct",
]
ara_d[protein]["Properties"] = {}
for k, v in zip(keys, results):
ara_d[protein]["Properties"][k] = v
return ara_d
def parse_nuc_sequence(self, n_seq, id=None, desc=None):
"""
Parses valid RNA sequence, translates nucleotides, calculates GC content and other methods available from ProteinAnalysis() in BioPython module.
Keyword arguments:
seq -- valid string sequence
id -- id obtained from FASTA file record (default None)
desc -- description obtained from FASTA file record (default None)
"""
try:
# append fasta sequence metadata
self.id.append(id)
self.description.append(desc)
self.nucleotide_sequence.append(n_seq)
# translate nucleotide string sequence
p_seq = self.translate_nucleotides(n_seq)
self.protein_sequence.append(p_seq)
# self.protein_sequence.append(str(record.seq.translate()).replace('*', ' '))
# GC content
self.gc_content.append(self.calculate_gc_content(n_seq))
# protein analysis methods
analysis = ProteinAnalysis(p_seq)
self.amino_acid_dict.append(analysis.get_amino_acids_percent())
self.molecular_weight.append(analysis.molecular_weight())
self.instability_index.append(analysis.instability_index())
self.aromaticity.append(analysis.aromaticity())
except Exception as e:
print('-'*80)
print(f"Exception in parsing uploaded virus sequence: {e}")
traceback.print_exc(file=sys.stdout)
print('-'*80)
def add_protein_characteristics(df):
df = df.copy()
aa_list = [
'A', 'C', 'E', 'D', 'G', 'F', 'I', 'H', 'K', 'M', 'L', 'N', 'Q', 'P',
'S', 'R', 'T', 'W', 'V', 'Y'
]
aa_dict = {}
for aa in aa_list:
aa_dict[aa] = []
prop_dict = {
'aromaticity': [],
'helix': [],
'turn': [],
'sheet': [],
'isoelectric_point': [],
'gravy': []
} #, 'flexibility': [], 'instability_index': []}
for i, s in enumerate(df['sequence']):
s = s.replace('B', 'D').replace('Z', 'E').replace('J', 'L').replace(
'X', 'G').replace('U', 'C').replace('O', 'K')
pa = ProteinAnalysis(s)
prop_dict['aromaticity'].append(pa.aromaticity())
prop_dict['isoelectric_point'].append(pa.isoelectric_point())
prop_dict['gravy'].append(pa.gravy())
# prop_dict['instability_index'].append(pa.instability_index())
# prop_dict['flexibility'].append(np.mean(pa.flexibility()))
for fraction, ss in zip(pa.secondary_structure_fraction(),
['helix', 'turn', 'sheet']):
prop_dict[ss].append(fraction)
for k, v in pa.get_amino_acids_percent().items():
aa_dict[k].append(v)
for k, v in aa_dict.items():
df[k] = v
for k, v in prop_dict.items():
df[k] = v
return df
isoelectricPt=[]
aromaticity=[]
aminoPercent=[]
secstruct=[]
hydrophob=[]
hydrophil=[]
surface=[]
gravy=[]
molweight=[]
instidx=[]
flex=[]
for seq in sequences:
X=ProteinAnalysis(str(seq))
isoelectricPt.append(X.isoelectric_point())
aromaticity.append(X.aromaticity())
aminoPercent.append(X.get_amino_acids_percent())
secstruct.append(X.secondary_structure_fraction())
# These features throw Key & Value Errors due to non standard amino acids
# (i.e. out of the 20 standard ones) e.g. X, U etc
try:
gravy.append(X.gravy())
molweight.append(X.molecular_weight())
instidx.append(X.instability_index())
flex.append(X.flexibility())
hydrophob.append(X.protein_scale(ProtParamData.kd, 9, 0.4))
hydrophil.append(X.protein_scale(ProtParamData.hw, 9, 0.4))
surface.append(X.protein_scale(ProtParamData.em, 9, 0.4))
except (KeyError,ValueError):
from Bio.SeqUtils.ProtParam import ProteinAnalysis
from Bio.SeqUtils import ProtParamData
from Bio import SeqIO
with open('../../samples/pdbaa') as fh:
for rec in SeqIO.parse(fh,'fasta'):
myprot = ProteinAnalysis(str(rec.seq))
print(myprot.count_amino_acids())
print(myprot.get_amino_acids_percent())
print(myprot.molecular_weight())
print(myprot.aromaticity())
print(myprot.instability_index())
print(myprot.flexibility())
print(myprot.isoelectric_point())
print(myprot.secondary_structure_fraction())
print(myprot.protein_scale(ProtParamData.kd, 9, .4))
total_area.append(list_areas[2])
print('done')
with temppathlib.TemporaryDirectory() as tmpdir:
# unzip the file with all the test PDBs
with zipfile.ZipFile(args.infile, "r") as zip_:
zip_.extractall(tmpdir.path)
for test_pdb in tmpdir.path.glob("*.pdb"):
for record in SeqIO.parse(test_pdb, "pdb-atom"):
sequence = str(record.seq).replace('X', 'G')
protein = ProteinAnalysis(str(sequence))
p_len.append(len(sequence))
mol_w.append(protein.molecular_weight())
iso_p.append(protein.isoelectric_point())
smell.append(protein.aromaticity())
taste_factor.append(protein.gravy())
insta_ind.append(protein.instability_index())
char_at_acid.append(protein.charge_at_pH(1))
char_at_neutral.append(protein.charge_at_pH(7))
char_at_base.append(protein.charge_at_pH(14))
helter_skeler.append(protein.secondary_structure_fraction()[0])
turnip.append(protein.secondary_structure_fraction()[1])
garfield.append(protein.secondary_structure_fraction()[2])
for x in amino_acids:
n = protein.count_amino_acids()[x]
for y in d_count.keys():
if y[-1] == x:
d_count[y].append(n)
for a in amino_acids:
m = protein.get_amino_acids_percent()[a]
#!/usr/bin/env python
import sys
from Bio import SeqIO
from Bio.SeqUtils.ProtParam import ProteinAnalysis
sys.stdout.write("ID\tMW\tIP\tgravy\tlength\tinstability\tmonoisotpoic\tSequence\n")
for record in SeqIO.parse(sys.stdin, "fasta"):
a = ProteinAnalysis(str(record.seq))
properties = list()
properties.append(record.id)
properties.append(a.molecular_weight())
properties.append(a.isoelectric_point())
properties.append(a.gravy())
properties.append(a.length)
properties.append(a.instability_index())
properties.append(a.aromaticity())
# always last column to make the output more readable
properties.append(a.sequence)
sys.stdout.write( '\t'.join(map(str, properties))+"\n" )
