def process_bam(insertion_type, bam, index):
if insertion_type.lower() == "melt":
input_list = melt_parser(index)
elif insertion_type.lower() == "tipseqhunter":
input_list = tipseq_parser(index)
else:
raise TypeError(
"{} is not recognized; Should be MELT or TIPseqHunter".format(
insertion_type))
logging.info("### Processing BAM File ###")
bam_handler = BamHandler(bam)
insertions = []
reads_dictionary = ReadsDict()
logging.info(" - Finding Reads in Target Regions")
for parsed_input in input_list:
current_insertion = Insertion(parsed_input)
reads_in_region = bam_handler.fetch_reads_in_region(current_insertion)
reads_dictionary += reads_in_region
current_insertion.read_keys_in_target_region = reads_in_region.keys()
insertions.append(current_insertion)
logging.info(" --- DONE ---")
logging.info(" - Finding Paired Ends")
longest_read = 0
for read in bam_handler.all_reads():
longest_read = max(read.query_length, longest_read)
if read.query_name in reads_dictionary:
reads_dictionary.insert(read)
logging.info(" --- DONE ---")
logging.info("### Processing Insertions ###")
insertions.sort(key=lambda x: x.chromosome)
return insertions, reads_dictionary, longest_read
def main(args):
group1 = args.group1
group2 = args.group2
sample_id = args.sample_id
bam_path = args.bam
insertion_list_path = args.index
me_ref_path = args.me_reference
host_ref_path = args.host_reference
genes_file_path = args.genes
keep_files = args.keep_files
# config = get_config_from_file()
# TODO - allow for multiple labels
# - eg : pos, unlabeled - pos - negative, pos
# TODO - make the reference subdirectories using the writer class
check_paths(
me_ref_path, host_ref_path, bam_path, insertion_list_path, genes_file_path,
)
output_folder_path = os.path.join(os.getcwd(), "output")
(reference_path, transposcope_path, track_path,) = create_output_folder_structure(
output_folder_path, group1, group2, sample_id
)
setup_logging()
logging.info("*** TranspoScope ***")
logging.info("### Input ###")
logging.info(" - Index File Path: {}".format(os.path.abspath(insertion_list_path)))
logging.info(" - BAM File Path: {}".format(os.path.abspath(bam_path)))
logging.info(
" - Mobile Element Reference File Path: {}".format(os.path.abspath(me_ref_path))
)
logging.info(
" - Host Genome Folder Path: {}".format(os.path.abspath(host_ref_path))
)
logging.info(
" - refFlat.txt Path: {}".format(
os.path.abspath(genes_file_path) if genes_file_path else "undefined"
)
)
logging.info(" - Group 1: {}".format(group1))
logging.info(" - Group 2: {}".format(group2))
logging.info(" - Keep Intermediate Files: {}".format(keep_files))
insertion_sites_reader = InsertionSiteReader(insertion_list_path)
logging.info("### Processing BAM File ###")
bam_handler = BamHandler(bam_path)
fasta_handler = FastaHandler(me_ref_path, host_ref_path)
insertions = []
reads_dictionary = {True: ReadsDict(), False: ReadsDict()}
logging.info(" - Finding Reads in Target Regions")
for insertion_stats in insertion_sites_reader.read_lines():
temp_insertion = Insertion(named_tuple=insertion_stats)
reads_in_region = bam_handler.fetch_reads_in_region(temp_insertion)
reads_dictionary[temp_insertion.THREE_PRIME] += reads_in_region
temp_insertion.read_keys_in_target_region = reads_in_region.keys()
insertions.append(temp_insertion)
logging.info(" --- DONE ---")
logging.info(" - Finding Paired Ends")
for read in bam_handler.all_reads():
if read.query_name in reads_dictionary[True]:
reads_dictionary[True].insert(read)
if read.query_name in reads_dictionary[False]:
reads_dictionary[False].insert(read)
logging.info(" --- DONE ---")
logging.info("### Processing Insertions ###")
file_writer = FileWriter()
realigner = Realigner(reference_path)
classifier = ReadClassifier(transposcope_path)
if genes_file_path:
gene_handler = GeneHandler(genes_file_path)
insertions.sort(key=lambda x: x.CHROMOSOME)
heading_table = {"Heading": ("ID", "Gene", "Probability"), "data": []}
ten_percent = len(insertions) / 10
next_log = ten_percent
completed = 0
for insertion in insertions:
file_name = "{i.CHROMOSOME}_{i.START}-{i.END}".format(i=insertion)
insertion.fasta_string = fasta_handler.generate_fasta_sequence(insertion)
fasta_path = file_writer.write_fasta(
reference_path,
file_name,
insertion.fasta_string,
">{i.CHROMOSOME}_{i.START}-{i.END}\n".format(i=insertion),
)
fastq1_path, fastq2_path = file_writer.write_fastq(
reference_path,
reads_dictionary[insertion.THREE_PRIME],
file_name,
insertion.read_keys_in_target_region,
)
sorted_bam_path = realigner.realign(
fasta_path, fastq1_path, fastq2_path, file_name
)
# TODO - add the option to not add gene information
if genes_file_path:
gene_info = gene_handler.find_nearest_gene(
insertion.CHROMOSOME, insertion.INSERTION_SITE
)
else:
gene_info = ("undefined", "rgb(3, 119, 190)")
# heading_table['Data']\
if insertion.THREE_PRIME:
end = "3"
else:
end = "5"
heading_table["data"].append(
[
"{}-{}({})".format(insertion.CHROMOSOME, insertion.CLIP_START, end),
gene_info,
"{:.2f}".format(insertion.PRED),
]
)
classifier.classify_insertion(insertion, sorted_bam_path)
completed += 1
if completed > next_log:
logging.info(
" - Percentage of insertions processed: {:.2%}.".format(
completed / len(insertions)
)
)
next_log += ten_percent
# TODO - write out index file
logging.info(" --- DONE ---")
if not keep_files:
logging.info("### Cleanup ###")
logging.info("Cleaning up generated files in {}".format(reference_path))
if os.path.exists(reference_path):
shutil.rmtree(os.path.dirname(reference_path))
logging.info(" --- DONE ---")
logging.info("### Building Bed File ###")
with open(os.path.join(track_path, "{}.bb".format(sample_id)), "w") as fh:
for insertion in insertions:
fh.write(
"{}\t{}\t{}\n".format(
insertion.CHROMOSOME, insertion.CLIP_START, insertion.CLIP_END,
)
)
logging.info(" --- DONE ---")
logging.info("### Building Website ###")
web_dir = os.path.join(os.getcwd(), "web")
logging.info(" The website is being built into: {}".format(web_dir))
file_writer.write_json(
os.path.join(transposcope_path, "table_info"), heading_table,
)
tree, found_table = build_tree(os.path.join(web_dir, "json"))
with open(os.path.join(web_dir, "manifest.json"), "w") as json_file:
json.dump(tree["json"], json_file)
logging.info(" --- DONE ---")
logging.info(
" to view the generated files in your browser run:\n\t\t'transposcope view {}'".format(
web_dir
)
)
def test_incremental_add(self):
reads_dict = ReadsDict()
reads_dict += {"a": 1}
reads_dict += {"b": 2, "c": 3, "d": 4}
self.assertDictEqual(reads_dict, {"a": 1, "b": 2, "c": 3, "d": 4})
def test_insert_read1_with_empty_list(self):
reads_dict = ReadsDict({"a": []})
self.assertRaises(ValueError, reads_dict.insert,
Read("a", True, False, 0))
def test_insert_read2_with_both_reads_present(self):
reads_dict = ReadsDict(
{"a": [Read("a", False, True, 0),
Read("a", True, False, 0)]})
result = reads_dict.insert(Read("a", False, True, 0))
self.assertIs(result, -1)
def test_insert_read1_with_read2_present(self):
reads_dict = ReadsDict({"a": [Read("a", True, False, 0)]})
result = reads_dict.insert(Read("a", False, True, 0))
self.assertIs(result, 0)
self.assertTrue(reads_dict["a"][0].is_read1)
self.assertTrue(reads_dict["a"][1].is_read2)
def test_reads_from_key_list(self):
reads_dict = ReadsDict()
reads_dict += {"a": 1, "b": 2, "c": 3, "d": 4}
iterator = reads_dict.reads_from_key_list(["b", "d"])
self.assertEqual(next(iterator), 2)
self.assertEqual(next(iterator), 4)