def get_frame(geneseq, gene_HXB2, genename, VERBOSE=0):
'''Get the frame by aligning the proteins'''
from seqanpy import align_local
from Bio.Seq import translate
from numpy import argmax
geneseq = ''.join(geneseq)
gene_HXB2 = ''.join(gene_HXB2)
if genename in ('tat1', 'rev1'):
gene_HXB2 = gene_HXB2[:len(gene_HXB2) - (len(gene_HXB2) % 3)]
elif genename in ('tat2', 'rev2'):
gene_HXB2 = gene_HXB2[len(gene_HXB2) % 3:]
prot_HXB2 = translate(gene_HXB2)
scores = []
for frame in xrange(3):
tmp = geneseq[frame:]
tmp = tmp[:len(tmp) - (len(tmp) % 3)]
tmp = translate(tmp)
(score, ali1, ali2) = align_local(prot_HXB2, tmp)
scores.append(score)
return argmax(scores)
def get_frame(geneseq, gene_HXB2, genename, VERBOSE=0):
'''Get the frame by aligning the proteins'''
from seqanpy import align_local
from Bio.Seq import translate
from numpy import argmax
geneseq = ''.join(geneseq)
gene_HXB2 = ''.join(gene_HXB2)
if genename in ('tat1', 'rev1'):
gene_HXB2 = gene_HXB2[:len(gene_HXB2) - (len(gene_HXB2) % 3)]
elif genename in ('tat2', 'rev2'):
gene_HXB2 = gene_HXB2[len(gene_HXB2) % 3:]
prot_HXB2 = translate(gene_HXB2)
scores = []
for frame in xrange(3):
tmp = geneseq[frame:]
tmp = tmp[:len(tmp) - (len(tmp) % 3)]
tmp = translate(tmp)
(score, ali1, ali2) = align_local(prot_HXB2, tmp)
scores.append(score)
return argmax(scores)
def align_dna(seqstr, refstr, require_full_cover=True):
if require_full_cover:
(score, alis, alir) = align_overlap(seqstr, refstr)
start = len(alir) - len(alir.lstrip('-'))
end = len(alir.rstrip('-'))
alist = alis[start:end]
alirt = alir[start:end]
else:
(score, alis, alir) = align_local(seqstr, refstr)
reftrim = alir.replace('-', '')
start = refstr.find(reftrim[:50])
end = refstr.rfind(reftrim[-50:]) + len(reftrim[-50:])
alist = ('N' * start) + alis + ('N' * (len(refstr) - end))
alirt = refstr[:start] + alir + refstr[end:]
return (alist, alirt)
def align_dna(seqstr, refstr, require_full_cover=True):
if require_full_cover:
(score, alis, alir) = align_overlap(seqstr, refstr)
start = len(alir) - len(alir.lstrip('-'))
end = len(alir.rstrip('-'))
alist = alis[start: end]
alirt = alir[start: end]
else:
(score, alis, alir) = align_local(seqstr, refstr)
reftrim = alir.replace('-', '')
start = refstr.find(reftrim[:50])
end = refstr.rfind(reftrim[-50:]) + len(reftrim[-50:])
alist = ('N' * start) + alis + ('N' * (len(refstr) - end))
alirt = refstr[:start] + alir + refstr[end:]
return (alist, alirt)
if VERBOSE >= 1:
print 'PCR', PCR,
if VERBOSE >= 2:
print ''
cons = build_consensus(bamfilename, len_reference, VERBOSE=VERBOSE,
block_len=block_len,
reads_per_alignment=n_reads_per_ali,
deltamax=deltamax)
consm = np.fromstring(cons, 'S1')
if VERBOSE >= 2:
print 'Reference length:', len_reference, 'consensus:', len(cons),
if len_reference != len(cons):
from seqanpy import align_local
(score, ali1, ali2) = align_local(''.join(refseq), cons)
alim1 = np.fromstring(ali1, 'S1')
alim2 = np.fromstring(ali2, 'S1')
n_diff = (alim1 != alim2).sum()
print 'overlap', len(alim1), 'n diffs.', n_diff
else:
n_diff = (refm != consm).sum()
print 'n diffs.', n_diff
if save_to_file:
if VERBOSE >= 2:
print 'Save to file'
fn_out = sample.get_consensus_filename(fragment, PCR=PCR)
consrec = SeqRecord(Seq(cons, ambiguous_dna),
id=samplename+'_consensus',
muts = mutations.loc[mutations['protein'] == protname]
# Get the structure
struc = pa.get_structure(protname, fdn + fns[protname])
# Get the right chain
scores = []
for chain in struc.get_chains():
# I is the compound (inhibitors) used for crystallization
if chain.id == 'I':
continue
#print('{:}, chain {:}'.format(protname, chain.id))
seql = [
d3to1.get(r.get_resname(), 'O') for r in chain.get_residues()
]
seq = ''.join(seql)
s, a1, a2 = align_local(prot['seq_aa'], seq)
scores.append(s)
chain = list(struc.get_chains())[np.argmax(s)]
seql = [d3to1.get(r.get_resname(), 'O') for r in chain.get_residues()]
seq = ''.join(seql)
# Flag all mutations
for m, mut in muts.iterrows():
mutations.at[m, 'PDB_fn'] = fns[protname]
mutations.at[m, 'PDB_id'] = fns[protname].split(
'_')[1].upper().split('.')[0]
mutations.at[m, 'PDB_chain'] = chain.id
s, a1, a2 = align_overlap(seq, mut['context_protein'])
# The focal allele is always small and is the only such letter
pos_in_context = 4
# Global pairwise alignment
seq1 = 'AAAGGTCTA'
seq2 = 'AAATCGA'
output = sap.align_global(seq1, seq2, band=5)
print output
# Overlap pairwise alignment
seq1 = 'AAAGGTCTA'
seq2 = 'ATCT'
output = sap.align_overlap(seq1, seq2)
print output
# Overlap pairwise alignment cutting flanks
seq1 = 'AAAGGTCTA'
seq2 = 'ATCT'
output = sap.align_overlap(seq1, seq2, cut_flanks=True)
print output
# Ladder pairwise alignment
seq1 = 'AAAGGTCTA'
seq2 = 'TCTAGGGAAACCC'
output = sap.align_ladder(seq1, seq2)
print output
# Local pairwise alignment
seq1 = 'AAAGGTCTACCGTAGCCT'
seq2 = 'AAGTCTAC'
output = sap.align_local(seq1, seq2)
print output
print ''
cons = build_consensus(bamfilename,
len_reference,
VERBOSE=VERBOSE,
block_len=block_len,
reads_per_alignment=n_reads_per_ali,
deltamax=deltamax)
consm = np.fromstring(cons, 'S1')
if VERBOSE >= 2:
print 'Reference length:', len_reference, 'consensus:', len(
cons),
if len_reference != len(cons):
from seqanpy import align_local
(score, ali1, ali2) = align_local(''.join(refseq), cons)
alim1 = np.fromstring(ali1, 'S1')
alim2 = np.fromstring(ali2, 'S1')
n_diff = (alim1 != alim2).sum()
print 'overlap', len(alim1), 'n diffs.', n_diff
else:
n_diff = (refm != consm).sum()
print 'n diffs.', n_diff
if save_to_file:
if VERBOSE >= 2:
print 'Save to file'
fn_out = sample.get_consensus_filename(fragment, PCR=PCR)
consrec = SeqRecord(
Seq(cons, ambiguous_dna),