def run_group_alignment (sequence_group):
seqrecords = []
for seq_id in sequence_group:
seqrecords.append(Bio.SeqRecord.SeqRecord (Bio.Seq.Seq(sequence_group[seq_id].replace ('-',''), Bio.Alphabet.IUPAC.IUPACAmbiguousDNA), id = seq_id, name = seq_id, description = ''))
if len (seqrecords) == 1:
refseq = seqrecords[0].format ('fasta')
return {'ref': refseq, 'alignment': refseq, 'seqs': seqrecords}
# find the longest sequence
seq_lengths = [len(record.seq) for record in seqrecords]
refseq = seqrecords[seq_lengths.index(max(seq_lengths))]
#print ('Ref set to %s' % refseq.id)
sm = BLOSUM62.load()
msa, discarded = align_to_refseq(
refseq,
seqrecords,
score_matrix=sm,
codon=True,
expected_identity=0.6,
keep_insertions=False
)
string_buffer = io.StringIO ()
Bio.SeqIO.write (msa, string_buffer, "fasta")
all_lines = string_buffer.getvalue()
string_buffer.close()
return {'ref': refseq.format ('fasta'), 'alignment': all_lines, 'seqs': seqrecords}
def run_group_alignment (seqrecords, refseqs):
sm = BLOSUM62.load()
seq_scores = {'names':{}}
sequence_names = {}
for seq in seqrecords:
sequence_names[seq.id] = len (sequence_names)
seqrecords.insert (0, 0)
for type in refseqs:
seq_scores [type] = {}
seq_scores['names'][type] = []
for seq in refseqs[type]:
seqrecords[0] = seq
seq_scores['names'][type].append (seq.id)
msa, discarded = align_to_refseq(
seq,
seqrecords,
score_matrix=sm,
codon=True,
expected_identity=0.4,
keep_insertions=True,
quiet=True
)
if (discarded):
print (discarded)
max_score = msa[0].annotations['_pbpscore']
max_len = msa[0].annotations['_nbpidentical']
seq_scores[type][seq.id] = [None for idx in range (len (sequence_names))]
msa.pop (0)
for successful in msa:
idx = sequence_names [successful.id]
seq_scores[type][seq.id][idx] = (successful.annotations['_pbpscore']/max_score+successful.annotations['_nbpidentical']/max_len)*0.5
return seq_scores
def run_group_alignment(seqrecords, refseqs):
sm = BLOSUM62.load()
seq_scores = {'names': {}}
sequence_names = {}
for seq in seqrecords:
sequence_names[seq.id] = len(sequence_names)
seqrecords.insert(0, 0)
for type in refseqs:
seq_scores[type] = {}
seq_scores['names'][type] = []
for seq in refseqs[type]:
seqrecords[0] = seq
seq_scores['names'][type].append(seq.id)
msa, discarded = align_to_refseq(seq,
seqrecords,
score_matrix=sm,
codon=True,
expected_identity=0.4,
keep_insertions=True,
quiet=True)
if (discarded):
print(discarded)
max_score = msa[0].annotations['_pbpscore']
max_len = msa[0].annotations['_nbpidentical']
seq_scores[type][seq.id] = [
None for idx in range(len(sequence_names))
]
msa.pop(0)
for successful in msa:
idx = sequence_names[successful.id]
seq_scores[type][seq.id][idx] = (
successful.annotations['_pbpscore'] / max_score +
successful.annotations['_nbpidentical'] / max_len) * 0.5
return seq_scores
