def trim_nongenomic_polyA_from_end(mapping, region_fetcher):
''' If a mapping ends in a polyA stretch, soft clip from the first
nongenomic A onward.
'''
if sam.contains_indel_pysam(mapping) or mapping.is_unmapped:
return mapping
first_ref_index = None
if mapping.is_reverse:
bases_to_trim = 0
poly_T_end = find_poly_T(mapping.seq)
for read_index, ref_index in mapping.aligned_pairs[::-1]:
if read_index == None:
# indels are filtered out above, so this can only be
# a skip from splicing
continue
if read_index > poly_T_end:
first_ref_index = ref_index
continue
ref_base = region_fetcher(mapping.tid, ref_index, ref_index + 1)
if ref_base != 'T':
bases_to_trim = read_index + 1
break
else:
# first_ref_index needs to be set to the last position
# that passed that 'are you genomic?' test
first_ref_index = ref_index
else:
first_ref_index = mapping.pos
bases_to_trim = 0
poly_A_start = find_poly_A(mapping.seq)
for read_index, ref_index in mapping.aligned_pairs:
if read_index < poly_A_start:
continue
ref_base = region_fetcher(mapping.tid, ref_index, ref_index + 1)
if ref_base != 'A':
bases_to_trim = len(mapping.seq) - read_index
break
if first_ref_index == None:
print mapping
raise ValueError('first_ref_index not set')
if bases_to_trim > 0:
mapping.pos = first_ref_index
trimmed_length = len(mapping.seq) - bases_to_trim
soft_clipped_block = [(sam.BAM_CSOFT_CLIP, bases_to_trim)]
if mapping.is_reverse:
# Remove blocks from the beginning.
trimmed_cigar = sam.truncate_cigar_blocks_from_beginning(
mapping.cigar, trimmed_length)
updated_cigar = soft_clipped_block + trimmed_cigar
else:
# Remove blocks from the end.
trimmed_cigar = sam.truncate_cigar_blocks_up_to(
mapping.cigar, trimmed_length)
updated_cigar = trimmed_cigar + soft_clipped_block
mapping.cigar = updated_cigar
if mapping.tags:
# Clear the MD tag since the possible removal of bases to the
# alignment may have made it inaccurate.
# TODO: now have machinery to make it accurate.
filtered_tags = filter(lambda t: t[0] != 'MD', mapping.tags)
mapping.tags = filtered_tags
set_nongenomic_length(mapping, bases_to_trim)
return mapping
def trim_mismatches_from_start(mapping, region_fetcher, type_counts):
''' Remove all consecutive Q30+ mismatches from the beginning of alignments,
under the assumption that these represent untemplated additions during
reverse transcription.
Characterize the mismatches into type_counts.
'''
if sam.contains_indel_pysam(mapping) or mapping.is_unmapped:
set_nongenomic_length(mapping, 0)
return mapping
if mapping.is_reverse:
aligned_pairs = mapping.aligned_pairs[::-1]
index_lookup = utilities.base_to_complement_index
else:
aligned_pairs = mapping.aligned_pairs
index_lookup = utilities.base_to_index
decoded_qual = fastq.decode_sanger(mapping.qual)
bases_to_trim = 0
found_trim_point = False
first_ref_index = None
for read_index, ref_index in aligned_pairs:
if read_index == None:
# This shouldn't be able to be triggered since alignments
# containing indels are ruled out above.
continue
if mapping.is_reverse:
corrected_read_index = mapping.qlen - 1 - read_index
else:
corrected_read_index = read_index
ref_base = region_fetcher(mapping.tid, ref_index, ref_index + 1)
read_base = mapping.seq[read_index]
read_qual = decoded_qual[read_index]
coords = (
mapping.qlen,
corrected_read_index,
read_qual,
index_lookup[ref_base],
index_lookup[read_base],
)
type_counts[coords] += 1
if not found_trim_point:
if read_base != ref_base and read_qual >= 30:
bases_to_trim += 1
else:
first_ref_index = ref_index
found_trim_point = True
if first_ref_index == None:
raise ValueError('first_ref_index not set')
if bases_to_trim == 0:
trimmed_mapping = mapping
else:
trimmed_mapping = pysam.AlignedRead()
trimmed_mapping.qname = mapping.qname
trimmed_mapping.tid = mapping.tid
# first_ref_index has been set above to the be index of the
# reference base aligned to the first non-trimmed base in the
# read. If the mapping is forward, this will be the new pos.
# If the mapping is reverse, the pos won't change.
if mapping.is_reverse:
first_ref_index = mapping.pos
trimmed_mapping.pos = first_ref_index
trimmed_mapping.is_reverse = mapping.is_reverse
trimmed_mapping.is_secondary = mapping.is_secondary
trimmed_mapping.mapq = mapping.mapq
if mapping.is_reverse:
# bases_to_trim is never zero here, so there is no danger
# of minus zero
trimmed_slice = slice(None, -bases_to_trim)
else:
trimmed_slice = slice(bases_to_trim, None)
trimmed_mapping.seq = mapping.seq[trimmed_slice]
trimmed_mapping.qual = mapping.qual[trimmed_slice]
trimmed_mapping.rnext = -1
trimmed_mapping.pnext = -1
trimmed_length = len(mapping.seq) - bases_to_trim
if mapping.is_reverse:
# Remove blocks from the end
trimmed_cigar = sam.truncate_cigar_blocks_up_to(
mapping.cigar, trimmed_length)
else:
# Remove blocks from the beginning
trimmed_cigar = sam.truncate_cigar_blocks_from_beginning(
mapping.cigar, trimmed_length)
trimmed_mapping.cigar = trimmed_cigar
return trimmed_mapping
def trim_nongenomic_polyA_from_end(mapping, region_fetcher):
''' If a mapping ends in a polyA stretch, soft clip from the first
nongenomic A onward.
'''
if sam.contains_indel_pysam(mapping) or mapping.is_unmapped:
return mapping
first_ref_index = None
if mapping.is_reverse:
bases_to_trim = 0
poly_T_end = find_poly_T(mapping.seq)
for read_index, ref_index in mapping.aligned_pairs[::-1]:
if read_index == None:
# indels are filtered out above, so this can only be
# a skip from splicing
continue
if read_index > poly_T_end:
first_ref_index = ref_index
continue
ref_base = region_fetcher(mapping.tid, ref_index, ref_index + 1)
if ref_base != 'T':
bases_to_trim = read_index + 1
break
else:
# first_ref_index needs to be set to the last position
# that passed that 'are you genomic?' test
first_ref_index = ref_index
else:
first_ref_index = mapping.pos
bases_to_trim = 0
poly_A_start = find_poly_A(mapping.seq)
for read_index, ref_index in mapping.aligned_pairs:
if read_index < poly_A_start:
continue
ref_base = region_fetcher(mapping.tid, ref_index, ref_index + 1)
if ref_base != 'A':
bases_to_trim = len(mapping.seq) - read_index
break
if first_ref_index == None:
print mapping
raise ValueError('first_ref_index not set')
if bases_to_trim > 0:
mapping.pos = first_ref_index
trimmed_length = len(mapping.seq) - bases_to_trim
soft_clipped_block = [(sam.BAM_CSOFT_CLIP, bases_to_trim)]
if mapping.is_reverse:
# Remove blocks from the beginning.
trimmed_cigar = sam.truncate_cigar_blocks_from_beginning(mapping.cigar, trimmed_length)
updated_cigar = soft_clipped_block + trimmed_cigar
else:
# Remove blocks from the end.
trimmed_cigar = sam.truncate_cigar_blocks_up_to(mapping.cigar, trimmed_length)
updated_cigar = trimmed_cigar + soft_clipped_block
mapping.cigar = updated_cigar
if mapping.tags:
# Clear the MD tag since the possible removal of bases to the
# alignment may have made it inaccurate.
# TODO: now have machinery to make it accurate.
filtered_tags = filter(lambda t: t[0] != 'MD', mapping.tags)
mapping.tags = filtered_tags
set_nongenomic_length(mapping, bases_to_trim)
return mapping
def filter_mappings(
mappings,
minimum_mapq=42,
max_insert_length=1000,
counts_dict=None,
verbose=False,
unmapped_fns=None,
):
''' Filters out unmapped, nonuniquely mapped, or discordantly mapped
reads.
'''
pair_counts = {
'total': 0,
'unmapped': 0,
'indel': 0,
'nonunique': 0,
'discordant': 0,
'disoriented': 0,
'unique': Counter(),
'mapqs': Counter(),
'fragment_lengths': Counter(),
'tids': Counter(),
}
if unmapped_fns:
R1_unmapped_fn, R2_unmapped_fn = unmapped_fns
R1_unmapped_fh = open(R1_unmapped_fn, 'w')
R2_unmapped_fh = open(R2_unmapped_fn, 'w')
for _, aligned_pair in utilities.group_by(mappings, key=lambda m: m.qname):
if len(aligned_pair) != 2:
raise ValueError(len(aligned_pair))
pair_counts['total'] += 1
R1_aligned, R2_aligned = aligned_pair
# If R2 is mapped but R1 isn't, R2 gets reported first.
if not R1_aligned.is_read1:
R1_aligned, R2_aligned = R2_aligned, R1_aligned
if (not R1_aligned.is_read1) or (not R2_aligned.is_read2):
raise ValueError(R1_aligned, R2_aligned)
pair_counts['mapqs'][R1_aligned.mapq] += 1
pair_counts['mapqs'][R2_aligned.mapq] += 1
if R1_aligned.is_unmapped or R2_aligned.is_unmapped:
pair_counts['unmapped'] += 1
if verbose:
logging.info('{0} was unmapped'.format(R1_aligned.qname))
if unmapped_fns:
R1_read = sam.mapping_to_Read(R1_aligned)
R2_read = sam.mapping_to_Read(R2_aligned)
R1_unmapped_fh.write(str(R1_read))
R2_unmapped_fh.write(str(R2_read))
elif is_discordant(R1_aligned, R2_aligned, max_insert_length):
pair_counts['discordant'] += 1
else:
pair_counts['tids'][R1_aligned.tid] += 1
if is_disoriented(R1_aligned, R2_aligned):
pair_counts['disoriented'] += 1
elif R1_aligned.mapq < minimum_mapq or R2_aligned.mapq < minimum_mapq:
pair_counts['nonunique'] += 1
if verbose:
logging.info('{0} was nonunique, {1}, {2}'.format(
R1_aligned.qname, R1_aligned.mapq, R2_aligned.mapq))
else:
pair_counts['unique'][R1_aligned.tid] += 1
fragment_length = abs(R1_aligned.tlen)
pair_counts['fragment_lengths'][fragment_length] += 1
if sam.contains_indel_pysam(
R1_aligned) or sam.contains_indel_pysam(R2_aligned):
pair_counts['indel'] += 1
yield R1_aligned, R2_aligned
if counts_dict != None:
counts_dict.update(pair_counts)
def trim_mismatches_from_start(mapping, region_fetcher, type_counts):
''' Remove all consecutive Q30+ mismatches from the beginning of alignments,
under the assumption that these represent untemplated additions during
reverse transcription.
Characterize the mismatches into type_counts.
'''
if sam.contains_indel_pysam(mapping) or mapping.is_unmapped:
set_nongenomic_length(mapping, 0)
return mapping
if mapping.is_reverse:
aligned_pairs = mapping.aligned_pairs[::-1]
index_lookup = utilities.base_to_complement_index
else:
aligned_pairs = mapping.aligned_pairs
index_lookup = utilities.base_to_index
decoded_qual = fastq.decode_sanger(mapping.qual)
bases_to_trim = 0
found_trim_point = False
first_ref_index = None
for read_index, ref_index in aligned_pairs:
if read_index == None:
# This shouldn't be able to be triggered since alignments
# containing indels are ruled out above.
continue
if mapping.is_reverse:
corrected_read_index = mapping.qlen - 1 - read_index
else:
corrected_read_index = read_index
ref_base = region_fetcher(mapping.tid, ref_index, ref_index + 1)
read_base = mapping.seq[read_index]
read_qual = decoded_qual[read_index]
coords = (mapping.qlen,
corrected_read_index,
read_qual,
index_lookup[ref_base],
index_lookup[read_base],
)
type_counts[coords] += 1
if not found_trim_point:
if read_base != ref_base and read_qual >= 30:
bases_to_trim += 1
else:
first_ref_index = ref_index
found_trim_point = True
if first_ref_index == None:
raise ValueError('first_ref_index not set')
if bases_to_trim == 0:
trimmed_mapping = mapping
else:
trimmed_mapping = pysam.AlignedRead()
trimmed_mapping.qname = mapping.qname
trimmed_mapping.tid = mapping.tid
# first_ref_index has been set above to the be index of the
# reference base aligned to the first non-trimmed base in the
# read. If the mapping is forward, this will be the new pos.
# If the mapping is reverse, the pos won't change.
if mapping.is_reverse:
first_ref_index = mapping.pos
trimmed_mapping.pos = first_ref_index
trimmed_mapping.is_reverse = mapping.is_reverse
trimmed_mapping.is_secondary = mapping.is_secondary
trimmed_mapping.mapq = mapping.mapq
if mapping.is_reverse:
# bases_to_trim is never zero here, so there is no danger
# of minus zero
trimmed_slice = slice(None, -bases_to_trim)
else:
trimmed_slice = slice(bases_to_trim, None)
trimmed_mapping.seq = mapping.seq[trimmed_slice]
trimmed_mapping.qual = mapping.qual[trimmed_slice]
trimmed_mapping.rnext = -1
trimmed_mapping.pnext = -1
trimmed_length = len(mapping.seq) - bases_to_trim
if mapping.is_reverse:
# Remove blocks from the end
trimmed_cigar = sam.truncate_cigar_blocks_up_to(mapping.cigar, trimmed_length)
else:
# Remove blocks from the beginning
trimmed_cigar = sam.truncate_cigar_blocks_from_beginning(mapping.cigar, trimmed_length)
trimmed_mapping.cigar = trimmed_cigar
return trimmed_mapping