def get_inds(self, seq):
lcas = motif_seq.LcSeq(seq, self.k, self.lca, 'lca')
lces = motif_seq.LcSeq(seq, self.k, self.lce, 'lce')
lca_in, lca_out = lcas._get_motif_indexes()
lce_in, lce_out = lces._get_motif_indexes()
ind_in = lca_in.union(lce_in)
return ind_in
def in_out_kappa(self):
df = pd.read_csv(self.train_fpi, sep='\t', index_col=0)
df = df[df['y'] == 0]
seqs = list(df['Sequence'])
all_deltas = []
net_charges = []
frac_charges = []
for seq in seqs:
ms = motif_seq.LcSeq(seq, self.k, self.lca, 'lca')
in_seq, out_seq = ms.seq_in_motif()
in_kmer, out_kmer = ms.overlapping_kmer_in_motif()
if len(in_kmer) > 20:
ka = kappa.KappaKmers(out_kmer, out_seq)
if ka.FCR() > 0.1:
delta = ka.deltaForm()
net_charges.append(ka.NCPR())
print(out_seq)
print(delta)
all_deltas.append(delta)
frac_charges.append(ka.FCR())
#plt.hist(net_charges)
plt.scatter(net_charges, all_deltas, alpha=0.5, color='grey')
#plt.ylim([0, 0.35])
plt.ylim([0, 0.5])
plt.xlim([-0.8, 0.8])
#plt.xlim([0, 0.4])
plt.xlabel('Net charge per residue', size=14)
plt.ylabel('Charge Asymmetry (Delta)', size=14)
plt.title('Outside LC Motifs')
plt.show()
def normal_charge_properties(self):
df = pd.read_csv(self.train_fpi, sep='\t', index_col=0)
df = df[df['y'] == 0]
seqs = list(df['Sequence'])
all_deltas = []
net_charges = []
frac_charges = []
all_seq_in = ''
for seq in seqs:
ms = motif_seq.LcSeq(seq, self.k, self.lca, 'lca')
in_seq, out_seq = ms.seq_in_motif()
in_kmer, out_kmer = ms.overlapping_kmer_in_motif()
if len(in_kmer) > 20:
ka = kappa.KappaKmers(out_kmer, out_seq)
delta = ka.deltaForm()
if ka.NCPR() > -0.1 and ka.NCPR() < 0.1:
if delta < 0.1:
ns = norm_score.NormScore()
score = ns.lc_norm_score([seq])[0]
if score > 20:
if ka.FCR() < 0.2:
all_seq_in += in_seq
analysed_seq = ProteinAnalysis(all_seq_in)
aa_perc = analysed_seq.get_amino_acids_percent()
print(aa_perc)
def in_out_kappa(self):
df = pd.read_csv(self.train_fpi, sep='\t', index_col=0)
df = df[df['y'] == 0]
seqs = list(df['Sequence'])
for seq in seqs:
ms = motif_seq.LcSeq(seq, self.k, self.lca, 'lca')
in_seq, out_seq = ms.seq_in_motif()
SeqOb = SequenceParameters(in_seq)
print(SeqOb.get_kappa())
seqOb = SequenceParameters(out_seq)
print(seqOb.get_kappa())
print('')