def get_sample_kmers(self, sample):
"""Get all kmers from a sample, either from the reference sequence, all possible kmers from an alphabet or
all kmers that cover a modified nucelotide
:param sample: AbstractSamples object
:return: set of desired kmers
"""
kmers = set()
# if motifs is present, process for all motifs with modified base
if sample.motifs:
for motif in sample.motifs:
kmers |= get_motif_kmers(motif,
self.k,
alphabet=self.canonical)
# if we want to limit kmers which were seen in reference sequence
if sample.kmers_from_reference:
for _, _, sequence in read_fasta(sample.bwa_reference):
kmers |= get_sequence_kmers(sequence, k=self.k, rev_comp=True)
else:
kmers |= {
x
for x in all_string_permutations(self.canonical, length=self.k)
}
return kmers
def make_positions_file(reference, output_path, motifs, overlap=False):
"""Creates a tsv file with the following format ("contig", "position", "strand", "change_from", "change_to").
Given a reference sequence and sets of sequence motif changes we report the location of each change.
NOTE: the motifs cannot create new characters on the opposite strand!
:param reference: path to reference sequence
:param output_path: output path of positions file
:param motifs: list of lists of find replace motifs ex: [("CCAGG","CFAGG"), ("CCTGG","CFTGG")]
:param overlap: if the motif can overlap with its self, find index of overlap if set to true
"""
rev_motifs = []
for motif in motifs:
rev_motifs.append([x[::-1] for x in motif])
with open(output_path, "w") as outfile:
for header, comment, sequence in read_fasta(reference):
fwd_seq = sequence
bwd_seq = reverse_complement(fwd_seq,
reverse=False,
complement=True).upper()
for index, old_char, substitution_char in find_motifs_sequence_positions(
fwd_seq, motifs, overlap=overlap):
outfile.write(header + "\t" + np.str(index) + "\t" + "+" +
"\t" + old_char + "\t" + substitution_char +
"\n")
for index, old_char, substitution_char in find_motifs_sequence_positions(
bwd_seq, rev_motifs, overlap=overlap):
outfile.write(header + "\t" + np.str(index) + "\t" + "-" +
"\t" + old_char + "\t" + substitution_char +
"\n")
return output_path
def replace_motif_reference_positions(reference_location,
sub_fasta_path,
motifs,
overlap=False):
"""Replace motif reference sequence to a specific path
:param reference_location: input reference
:param sub_fasta_path: location of edited reference
:param motifs: list of motif's which need to be replaced: eg [[find, replace]], [["CCAGG", "CEAGG"]]
:param overlap: of overlap is possible, replace with overlap: eg [["AAA", "AAT"]] : AAAA -> AATT
"""
if os.path.isfile(sub_fasta_path):
print(
"[substitute_reference_positions] Substituted reference fasta file exists: {}"
.format(sub_fasta_path))
return sub_fasta_path
else:
print(
"[substitute_reference_positions] Creating substituted reference fasta file: {}"
.format(sub_fasta_path))
# write
with open(sub_fasta_path, 'w') as outfasta:
for header, comment, sequence in read_fasta(reference_location):
subst_sequence = replace_motifs_sequence_positions(
sequence, motifs, overlap)
print(
">%s %s\n%s" %
(header, "substituted:{}".format(motifs), subst_sequence),
file=outfasta)
return sub_fasta_path
def replace_periodic_reference_positions(reference_location,
sub_fasta_path,
step,
offset,
substitution_char='X'):
"""Edit and write a reference sequence to a specified path by replacing periodic characters
note: if sub_fasta_path exists it will return the path without creating a new file
:param reference_location: input reference
:param sub_fasta_path: location of edited reference
:param step: size of gap between substitution characters
:param offset: offset of when to start creating substiutions
:param substitution_char: character to replace original character
"""
if os.path.isfile(sub_fasta_path):
print(
"[substitute_reference_positions] Substituted reference fasta file exists: {}"
.format(sub_fasta_path))
return sub_fasta_path
else:
print(
"[substitute_reference_positions] Creating substituted reference fasta file: {}"
.format(sub_fasta_path))
# write
with open(sub_fasta_path, 'w') as outfasta:
for header, comment, sequence in read_fasta(reference_location):
subst_sequence = replace_periodic_sequence_positions(
sequence, step, offset, substitution_char)
print(">%s %s\n%s" %
(header, "substituted:{},step:{},offset:{}".format(
substitution_char, step, offset), subst_sequence),
file=outfasta)
return sub_fasta_path
def get_first_seq(fasta_file):
assert os.path.isfile(fasta_file)
i = 0
seq = ""
for t, c, s in read_fasta(fasta_file):
seq += s
i += 1
if i > 0:
break
return seq
def get_sequence(path_to_fasta):
seqs = []
for header, comment, sequence in read_fasta(path_to_fasta):
seqs.append(sequence)
assert len(seqs) > 0, "ERROR parsing sequence {}".format(len(seqs))
if len(seqs) > 1:
print("Taking first sequence of {}".format(len(seqs)))
return seqs[0]
def get_reference_sequence(path_to_fasta):
seqs = []
for header, comment, sequence in read_fasta(path_to_fasta):
seqs.append(sequence)
assert len(seqs) > 0, "Didn't find any sequences in the reference file"
if len(seqs) > 1:
print("[NOTICE] Found more than one sequence in the reference file, using the first one")
return seqs[0]
def processReferenceFasta(fasta,
work_folder,
motif_key=None,
sub_char=None,
positions_file=None):
"""loops over all of the contigs in the reference file, writes the forward and backward sequences
as flat files (no headers or anything) for signalMachine, returns a dict that has the sequence
names as keys and the paths to the processed sequence as keys
"""
if positions_file is not None and motif_key is not None:
raise RuntimeError(
"[processReferenceFasta]Cannot specify motif key and ambiguity position file"
)
if positions_file is not None and sub_char is not None:
raise RuntimeError(
"[processReferenceFasta]Cannot specify a substitution character and a ambiguity position file"
)
if positions_file is not None:
if not os.path.exists(positions_file):
raise RuntimeError(
"[processReferenceFasta]Did not find ambiguity position file here: %s"
% positions_file)
positions = CustomAmbiguityPositions(positions_file)
else:
positions = None
if motif_key is None and sub_char is None and positions_file is None:
return fasta, None
else:
fw_fasta_path = work_folder.add_file_path("forward.{}".format(
os.path.basename(fasta)))
bw_fasta_path = work_folder.add_file_path("backward.{}".format(
os.path.basename(fasta)))
if not os.path.exists(fw_fasta_path) and not os.path.exists(
bw_fasta_path):
print(
"[SignalALignment.run]NOTICE: Creating forward and backward fasta files."
)
with open(bw_fasta_path,
'w') as bw_outfasta, open(fw_fasta_path,
'w') as fw_outfasta:
for header, comment, sequence in read_fasta(fasta):
# the motif label allows us to make multiple copies of the reference with unique file names
# motif_lab = "" if motif_key is None else "%s." % motif_key
# these are the paths to the flat files that have the references
# signalAlign likes uppercase
if motif_key is not None:
motif, ok = getMotif(motif_key, sequence)
if not ok:
raise RuntimeError(
"[processReferenceFasta]Illegal motif key %s" %
motif_key)
fw_sequence = motif.forwardSubstitutedSequence(
sub_char)
bw_sequence = motif.complementSubstitutedSequence(
sub_char)
elif positions is not None:
fw_sequence = positions.getForwardSequence(
contig=header, raw_sequence=sequence.upper())
bw_sequence = positions.getBackwardSequence(
contig=header, raw_sequence=sequence.upper())
else:
fw_sequence = sequence.upper()
bw_sequence = reverse_complement(fw_sequence,
reverse=False,
complement=True)
print(">%s %s\n%s" % (header, "backward", bw_sequence),
file=bw_outfasta)
print(">%s %s\n%s" % (header, "forward", fw_sequence),
file=fw_outfasta)
return fw_fasta_path, bw_fasta_path
def processReferenceFasta(fasta,
work_folder,
motif_key=None,
sub_char=None,
positions_file=None):
"""loops over all of the contigs in the reference file, writes the forward and backward sequences
as flat files (no headers or anything) for signalMachine, returns a dict that has the sequence
names as keys and the paths to the processed sequence as keys
"""
if positions_file is not None and motif_key is not None:
raise RuntimeError(
"[processReferenceFasta]Cannot specify motif key and ambiguity position file"
)
if positions_file is not None and sub_char is not None:
raise RuntimeError(
"[processReferenceFasta]Cannot specify a substitution character and a ambiguity position file"
)
if positions_file is not None:
if not os.path.exists(positions_file):
raise RuntimeError(
"[processReferenceFasta]Did not find ambiguity position file here: %s"
% positions_file)
positions = CustomAmbiguityPositions(positions_file)
else:
positions = None
ref_sequence_map = {}
for header, comment, sequence in read_fasta(fasta):
# the motif label allows us to make multiple copies of the reference with unique file names
motif_lab = "" if motif_key is None else "%s." % motif_key
# these are the paths to the flat files that have the references
fw_path = work_folder.add_file_path("%s%s.%s.forward.txt" %
(motif_lab, header, sub_char))
bw_path = work_folder.add_file_path("%s%s.%s.backward.txt" %
(motif_lab, header, sub_char))
# signalAlign likes uppercase
if motif_key is not None:
motif, ok = getMotif(motif_key, sequence)
if not ok:
raise RuntimeError(
"[processReferenceFasta]Illegal motif key %s" % motif_key)
fw_sequence = motif.forwardSubstitutedSequence(sub_char)
bw_sequence = motif.complementSubstitutedSequence(sub_char)
elif positions is not None:
fw_sequence = positions.getForwardSequence(
contig=header, raw_sequence=sequence.upper())
bw_sequence = positions.getBackwardSequence(
contig=header, raw_sequence=sequence.upper())
else:
fw_sequence = sequence.upper()
bw_sequence = reverse_complement(fw_sequence,
reverse=False,
complement=True)
with open(fw_path, 'w') as fH:
fH.write("%s\n" % fw_sequence)
with open(bw_path, 'w') as fH:
fH.write("%s\n" % bw_sequence)
ref_sequence_map[header] = {"forward": fw_path, "backward": bw_path}
return ref_sequence_map
def processReferenceFasta(fasta,
work_folder,
name,
motifs=None,
positions_file=None):
"""loops over all of the contigs in the reference file, writes the forward and backward sequences
as flat files (no headers or anything) for signalMachine, returns a dict that has the sequence
names as keys and the paths to the processed sequence as keys
:param fasta: path to un-edited fasta file
:param work_folder: FolderHandler object
:param motifs: list of tuple pairs for motif edits. ex [["CCAGG", "CEAGG"]]
:param positions_file: ambiguous positions file which can be processed via CustomAmbiguityPositions
:return: paths to possibly edited forward reference sequence and backward reference sequence
"""
positions = None
# if no processing needs to happen
if positions_file is None and motifs is None:
return fasta, None
# Cant pass positions file and motifs
if positions_file is not None and motifs is not None:
raise RuntimeError(
"[processReferenceFasta] Cannot specify motif key and ambiguity position file"
)
# get positions object (if appropriate)
if positions_file:
if not os.path.exists(positions_file):
raise RuntimeError(
"[processReferenceFasta] Did not find ambiguity position file here: %s"
% positions_file)
positions = CustomAmbiguityPositions(positions_file)
# process fasta
fw_fasta_path = work_folder.add_file_path("forward.{}.{}".format(
name, os.path.basename(fasta)))
bw_fasta_path = work_folder.add_file_path("backward.{}.{}".format(
name, os.path.basename(fasta)))
print(
"[SignalAlignment.run] NOTICE: Creating forward and backward fasta files."
)
with open(bw_fasta_path, 'w') as bw_outfasta, open(fw_fasta_path,
'w') as fw_outfasta:
for header, comment, sequence in read_fasta(fasta):
# signalAlign likes uppercase
if positions is not None:
fw_sequence = positions.getForwardSequence(
contig=header, raw_sequence=sequence.upper())
bw_sequence = positions.getBackwardSequence(
contig=header, raw_sequence=sequence.upper())
else:
fw_sequence = sequence.upper()
bw_sequence = reverse_complement(fw_sequence,
reverse=False,
complement=True).upper()
if motifs:
fw_sequence = replace_motifs_sequence_positions(
fw_sequence, motifs, True)
bw_sequence = replace_motifs_sequence_positions(
bw_sequence, motifs, True)
print(">%s %s\n%s" % (header, "backward", bw_sequence),
file=bw_outfasta)
print(">%s %s\n%s" % (header, "forward", fw_sequence),
file=fw_outfasta)
return fw_fasta_path, bw_fasta_path
