def aa_analysis(df, property):
if property == "ncpr":
df = df[pd.notnull(df['Amino_acids'])]
df[["AA1","AA2"]] = df['Amino_acids'].str.split('/',expand=True)
isoelectric_point = []
for sequence in df["AA1"]:
try:
cdr3 = ProteinAnalysis(str(sequence))
cidercdr3 = SequenceParameters(str(sequence))
isoelectric_point.append(cidercdr3.get_NCPR())
except:
isoelectric_point.append(0)
pass
df["AA1_Iso"] = isoelectric_point
isoelectric_point2 = []
for sequence in df["AA2"]:
try:
cdr3 = ProteinAnalysis(str(sequence))
cidercdr3 = SequenceParameters(str(sequence))
isoelectric_point2.append(cidercdr3.get_NCPR())
except:
isoelectric_point2.append(0)
pass
df["AA2_Iso"] = isoelectric_point2
df["AA_Iso_Delta"] = df["AA2_Iso"] - df["AA1_Iso"]
df = df[["AA1_Iso", "AA2_Iso", "AA_Iso_Delta"]]
elif property == "uversky_hydropathy":
df = df[pd.notnull(df['Amino_acids'])]
df[["AA1","AA2"]] = df['Amino_acids'].str.split('/',expand=True)
isoelectric_point = []
for sequence in df["AA1"]:
try:
cdr3 = ProteinAnalysis(str(sequence))
cidercdr3 = SequenceParameters(str(sequence))
isoelectric_point.append(cidercdr3.get_uversky_hydropathy())
except:
isoelectric_point.append(0)
pass
df["AA1_Iso"] = isoelectric_point
isoelectric_point2 = []
for sequence in df["AA2"]:
try:
cdr3 = ProteinAnalysis(str(sequence))
cidercdr3 = SequenceParameters(str(sequence))
isoelectric_point2.append(cidercdr3.get_uversky_hydropathy())
except:
isoelectric_point2.append(0)
pass
df["AA2_Iso"] = isoelectric_point2
df["AA_Iso_Delta"] = df["AA2_Iso"] - df["AA1_Iso"]
df = df[["AA1_Iso", "AA2_Iso", "AA_Iso_Delta"]]
return df
def get_features_physchem(seq):
"""Return dictionary of all features associated with physiochemical properties."""
SeqOb = SequenceParameters(seq)
return {'fraction_acidic': fraction_acidic(seq), 'fraction_basic': fraction_basic(seq),
'fraction_aliphatic': fraction_aliphatic(seq), 'fraction_aromatic': fraction_aromatic(seq),
'fraction_polar': fraction_polar(seq), 'fraction_disorder': fraction_disorder(seq), 'fraction_chainexp': fraction_chainexp(seq),
'hydropathy': SeqOb.get_uversky_hydropathy(), 'isopoint': predict_isoelectric_point(seq),
'loglen': log2(len(seq)), 'PPII_propensity': SeqOb.get_PPII_propensity()}
def feat_physchem(seq):
SeqOb = SequenceParameters(seq)
return {
'frac_acidic': frac_acidic(seq),
'frac_basic': frac_basic(seq),
'frac_aliphatic': frac_aliphatic(seq),
'frac_chainexp': frac_chainexp(seq),
'frac_polar': frac_polar(seq),
'frac_aromatic': frac_aromatic(seq),
'frac_disorder': frac_disorder(seq),
'loglen': log2(len(seq)),
'hydropathy': SeqOb.get_uversky_hydropathy(),
'iso_point': ProteinAnalysis(seq).isoelectric_point(),
'PPII_prop': SeqOb.get_PPII_propensity()
}
