def ribo_seq_read_counting_raw(gene,
sqlite_path_organism,
sqlite_path_reads,
count_type="range",
unique=True):
supported = get_protein_coding_transcript_ids(gene, sqlite_path_organism)
exclude = True
orf_regions = genomic_orf_coordinate_ranges(gene, sqlite_path_organism,
supported)
if unique:
orf_regions = get_unique_regions(orf_regions)
# unique_regions = get_unique_regions(genomic_exon_coordinates)
# all_junctions, unique_junctions = unique_exon_junctions(genomic_exon_coordinates)
# junction_scores = get_scores_per_exonjunction_for_gene(gene, sqlite_path_organism, sqlite_path_reads, supported, filter=exclude)
# transcripts_to_explain_reads = explain_exon_junctions(junction_scores, all_junctions, unique_junctions)
try:
["range", "fiveprime", "asite"].index(count_type)
except ValueError:
print(
"The count type must be one of 'range', 'fiveprime' or 'asite'. "
"count_type refers to the part of the read that is used in the feature counting process"
)
return "ERROR"
if count_type == "range":
genomic_read_ranges = get_read_ranges_genomic_location(
gene,
sqlite_path_reads,
sqlite_path_organism,
supported,
filter=exclude)
counts = count_readranges_supporting_exons_per_transcript(
orf_regions, genomic_read_ranges)
if count_type == "fiveprime":
genomic_read_positions = get_reads_per_genomic_location_fiveprime(
gene,
sqlite_path_reads,
sqlite_path_organism,
supported,
filter=exclude)
counts = count_read_supporting_regions_per_transcript(
orf_regions, genomic_read_positions)
if count_type == "asite":
genomic_read_positions = get_reads_per_genomic_location_asite(
gene,
sqlite_path_reads,
sqlite_path_organism,
supported,
filter=exclude)
counts = count_read_supporting_regions_per_transcript(
orf_regions, genomic_read_positions)
sum_of_exon_counts = {}
maximum_sum = 0
print(counts)
return counts
def ribo_seq_read_counting(gene, sqlite_path_organism, sqlite_path_reads, count_type="range", unique=True):
supported = get_protein_coding_transcript_ids(gene, sqlite_path_organism)
exclude = True
orf_regions = genomic_orf_coordinate_ranges(gene, sqlite_path_organism, supported)
if unique:
orf_regions = get_unique_regions(orf_regions)
# unique_regions = get_unique_regions(genomic_exon_coordinates)
# all_junctions, unique_junctions = unique_exon_junctions(genomic_exon_coordinates)
# junction_scores = get_scores_per_exonjunction_for_gene(gene, sqlite_path_organism, sqlite_path_reads, supported, filter=exclude)
# transcripts_to_explain_reads = explain_exon_junctions(junction_scores, all_junctions, unique_junctions)
try:
["range", "fiveprime", "asite"].index(count_type)
except ValueError:
print("The count type must be one of 'range', 'fiveprime' or 'asite'. "
"count_type refers to the part of the read that is used in the feature counting process")
return "ERROR"
if count_type == "range":
genomic_read_ranges = get_read_ranges_genomic_location(gene, sqlite_path_reads, sqlite_path_organism, supported,
filter=exclude)
counts = count_readranges_supporting_exons_per_transcript(orf_regions, genomic_read_ranges)
if count_type == "fiveprime":
genomic_read_positions = get_reads_per_genomic_location_fiveprime(gene, sqlite_path_reads, sqlite_path_organism,
supported, filter=exclude)
counts = count_read_supporting_regions_per_transcript(orf_regions, genomic_read_positions)
if count_type == "asite":
genomic_read_positions = get_reads_per_genomic_location_asite(gene, sqlite_path_reads, sqlite_path_organism,
supported, filter=exclude)
counts = count_read_supporting_regions_per_transcript(orf_regions, genomic_read_positions)
sum_of_exon_counts = {}
maximum_sum = 0
region_coverage = get_coverage_per_region(orf_regions, counts)
coverage = average_coverage_per_transcript(region_coverage)
# rankings = rank_based_on_dict_values(coverage)
# rankings_dict = {}
# for i in rankings:
# print i
# if i[0] not in rankings_dict:
# rankings_dict[i[0]] = i[1]
# for transcript in counts:
# transcript_sum = sum(counts[transcript])
# # if (transcript_sum > 0) and (transcript not in transcripts_to_explain_reads):
# # transcripts_to_explain_reads.append(str(transcript))
# sum_of_exon_counts[transcript] = transcript_sum
#
# if transcript_sum > maximum_sum:
# maximum_sum = transcript_sum
# maximum_transcript = transcript
# print "highest sum of counts: {maximum_transcript} with a total sum of: {maximum_sum}".format(maximum_transcript=str(maximum_transcript), maximum_sum=maximum_sum)
return coverage
def rna_seq_read_counting(gene, sqlite_path_organism, sqlite_path_reads, exclude=True, count_type="range"):
if exclude:
supported = filter_unsupported_transcripts(gene, sqlite_path_organism, sqlite_path_reads)
else:
gene_info = get_gene_info(gene, sqlite_path_organism)
supported = [i[0] for i in gene_info]
genomic_exon_coordinates = genomic_exon_coordinate_ranges(gene, sqlite_path_organism, supported)
unique_regions = get_unique_regions(genomic_exon_coordinates)
all_junctions, unique_junctions = unique_exon_junctions(genomic_exon_coordinates)
junction_scores = get_scores_per_exonjunction_for_gene(gene, sqlite_path_organism, sqlite_path_reads, supported, filter=exclude)
transcripts_to_explain_reads = explain_exon_junctions(junction_scores, all_junctions, unique_junctions)
try:
["range", "fiveprime", "asite"].index(count_type)
except ValueError:
print("The count type must be one of 'range', 'fiveprime' or 'asite'. "
"count_type refers to the part of the read that is used in the feature counting process")
return "ERROR"
if count_type == "range":
genomic_read_ranges = get_read_ranges_genomic_location(gene, sqlite_path_reads, sqlite_path_organism, supported,
filter=exclude)
counts = count_readranges_supporting_exons_per_transcript(unique_regions, genomic_read_ranges)
if count_type == "fiveprime":
genomic_read_positions = get_reads_per_genomic_location_fiveprime(gene, sqlite_path_reads, sqlite_path_organism,
supported, filter=exclude)
counts = count_read_supporting_regions_per_transcript(unique_regions, genomic_read_positions)
if count_type == "asite":
genomic_read_positions = get_reads_per_genomic_location_asite(gene, sqlite_path_reads, sqlite_path_organism,
supported, filter=exclude)
counts = count_read_supporting_regions_per_transcript(unique_regions, genomic_read_positions)
sum_of_exon_counts = {}
maximum_sum = 0
for transcript in counts:
transcript_sum = sum(counts[transcript])
if (transcript_sum > 0) and (transcript not in transcripts_to_explain_reads):
transcripts_to_explain_reads.append(str(transcript))
sum_of_exon_counts[transcript] = transcript_sum
if transcript_sum > maximum_sum:
maximum_sum = transcript_sum
maximum_transcript = transcript
print("The transcript with most uniquely mapped reads is {maximum_transcript} with a score of {maximum_sum}".format(maximum_transcript=str(maximum_transcript), maximum_sum=maximum_sum) )
return transcripts_to_explain_reads
def classify_regions_pos_neg(gene,
sqlite_path_reads,
sqlite_path_organism,
regions,
supported,
exclude=True,
count_type="range"):
# classify the genomic coordinate ranges into pos or neg. pos if 1 or more read supports it. Returns a nested dictionary
# with two nests. 1st on transcripts as keys, 2nd with pos or neg as keys
if count_type == "range":
genomic_read_ranges = get_read_ranges_genomic_location(
gene,
sqlite_path_reads,
sqlite_path_organism,
supported,
filter=exclude)
counts = count_readranges_supporting_exons_per_transcript(
regions, genomic_read_ranges)
if count_type == "fiveprime":
genomic_read_positions = get_reads_per_genomic_location_fiveprime(
gene,
sqlite_path_reads,
sqlite_path_organism,
supported,
filter=exclude)
counts = count_read_supporting_regions_per_transcript(
regions, genomic_read_positions)
if count_type == "asite":
genomic_read_positions = get_reads_per_genomic_location_asite(
gene,
sqlite_path_reads,
sqlite_path_organism,
supported,
filter=exclude)
counts = count_read_supporting_regions_per_transcript(
regions, genomic_read_positions)
classified = {}
for transcript in counts:
if transcript not in classified:
classified[transcript] = {'pos': [], 'neg': []}
for i in zip(regions[transcript], counts[transcript]):
if i[1] > 0:
classified[transcript]['pos'].append(i[0])
else:
classified[transcript]['neg'].append(i[0])
return classified