def test_database(self):
if find_cmd("sqlite3"):
with make_workdir() as workdir:
arg = namedtuple('args', 'debug print_debug json ')
args = arg(True, True, "../../data/examples/seqcluster.json")
initialize_logger(".", args.debug, args.print_debug)
logger = mylog.getLogger(__name__)
logger.info(args)
logger.info("Reading data")
data = load_data(args.json)
logger.info("Create database")
make_database(data)
def test_database(self):
if find_cmd("sqlite3"):
with make_workdir() as workdir:
arg = namedtuple('args', 'debug print_debug json ')
args = arg(True, True, "../../data/examples/seqcluster.json")
initialize_logger(".", args.debug, args.print_debug)
logger = mylog.getLogger(__name__)
logger.info(args)
logger.info("Reading data")
data = load_data(args.json)
logger.info("Create database")
make_database(data)
def test_databse(self):
if find_cmd("sqlite3"):
mod_dir = os.path.dirname(inspect.getfile(seqcluster)).replace("seqcluster/", "")
os.chdir(os.path.join(mod_dir, "data/examples"))
arg = namedtuple('args', 'debug print_debug json ')
args = arg(True, True, "seqcluster.json")
initialize_logger(".", args.debug, args.print_debug)
logger = mylog.getLogger(__name__)
logger.info(args)
logger.info("Reading data")
data = load_data(args.json)
logger.info("Create databse")
make_database(data)
def report(args):
"""
Create report in html format
"""
logger.info("reading sequeces")
data = load_data(args.json)
logger.info("create profile")
data = make_profile(data, os.path.join(args.out, "profiles"), args)
logger.info("create database")
make_database(data, "seqcluster.db", args.out)
logger.info("Done. Download https://github.com/lpantano/seqclusterViz/archive/master.zip to browse the output.")
def report(args):
"""
Create report in html format
"""
logger.info("Reading sequences")
data = load_data(args.json)
logger.info("Create profile")
data = make_profile(data, os.path.join(args.out, "profiles"), args)
logger.info("Create database")
make_database(data, "seqcluster.db", args.out)
logger.info("Done. Download https://github.com/lpantano/seqclusterViz/archive/master.zip to browse the output.")
def test_predict(self):
if find_cmd("tRNAscan-SE"):
mod_dir = os.path.dirname(inspect.getfile(seqcluster)).replace("seqcluster/", "")
os.chdir(os.path.join(mod_dir, "data/examples"))
out_dir = os.path.join(mod_dir, "data/examples/predictions")
arg = namedtuple('args', 'debug print_debug MIN_SEQ json reference')
args = arg(True, True, 1, "seqcluster.json", "../genomes/genome.fa")
initialize_logger(".", args.debug, args.print_debug)
logger = mylog.getLogger(__name__)
logger.info(args)
logger.info("Reading data")
data = load_data(args.json)
logger.info("Start prediction")
make_predictions(data, out_dir, args)
def test_databse(self):
if find_cmd("sqlite3"):
mod_dir = os.path.dirname(inspect.getfile(seqcluster)).replace(
"seqcluster/", "")
os.chdir(os.path.join(mod_dir, "data/examples"))
arg = namedtuple('args', 'debug print_debug json ')
args = arg(True, True, "seqcluster.json")
initialize_logger(".", args.debug, args.print_debug)
logger = mylog.getLogger(__name__)
logger.info(args)
logger.info("Reading data")
data = load_data(args.json)
logger.info("Create databse")
make_database(data)
def test_predict(self):
if find_cmd("tRNAscan-SE"):
mod_dir = os.path.dirname(inspect.getfile(seqcluster)).replace(
"seqcluster/", "")
os.chdir(os.path.join(mod_dir, "data/examples"))
out_dir = os.path.join(mod_dir,
"data/examples/test_out_predictions")
arg = namedtuple('args',
'debug print_debug MIN_SEQ json reference')
args = arg(True, True, 1, "seqcluster.json",
"../genomes/genome.fa")
initialize_logger(".", args.debug, args.print_debug)
logger = mylog.getLogger(__name__)
logger.info(args)
logger.info("Reading data")
data = load_data(args.json)
logger.info("Start prediction")
is_tRNA(data, out_dir, args)
def explore(args):
"""Create mapping of sequences of two clusters
"""
logger.info("reading sequeces")
data = load_data(args.json)
logger.info("get sequences from json")
#get_sequences_from_cluster()
c1, c2 = args.names.split(",")
seqs, names = get_sequences_from_cluster(c1, c2, data[0])
loci = get_precursors_from_cluster(c1, c2, data[0])
logger.info("map all sequences to all loci")
print("%s" % (loci))
map_to_precursors(seqs, names, loci, os.path.join(args.out, "map.tsv"), args)
#map_sequences_w_bowtie(sequences, precursors)
logger.info("plot sequences on loci")
#get_matrix_position()
#plot_sequences()
logger.info("Done")
def explore(args):
"""Create mapping of sequences of two clusters
"""
logger.info("Reading sequences")
data = load_data(args.json)
logger.info("Get sequences from json")
#get_sequences_from_cluster()
c1, c2 = args.names.split(",")
seqs, names = get_sequences_from_cluster(c1, c2, data[0])
loci = get_precursors_from_cluster(c1, c2, data[0])
logger.info("Map all sequences to all loci")
print("%s" % (loci))
map_to_precursors(seqs, names, loci, os.path.join(args.out, "map.tsv"), args)
#map_sequences_w_bowtie(sequences, precursors)
logger.info("Plot sequences on loci")
#get_matrix_position()
#plot_sequences()
logger.info("Done")
def predictions(args):
"""
Create predictions of clusters
"""
logger.info(args)
logger.info("Reading sequences")
out_file = os.path.abspath(os.path.splitext(args.json)[0] + "_prediction.json")
data = load_data(args.json)
out_dir = os.path.abspath(safe_dirs(os.path.join(args.out, "predictions")))
logger.info("Make predictions")
data = is_tRNA(data, out_dir, args)
if args.coral:
logger.info("Make CoRaL predictions")
run_coral(data, out_dir, args)
write_data(data[0], out_file)
logger.info("Done")
def predictions(args):
"""
Create predictions of clusters
"""
logger.info(args)
logger.info("reading sequeces")
out_file = os.path.abspath(os.path.splitext(args.json)[0] + "_prediction.json")
data = load_data(args.json)
out_dir = os.path.abspath(safe_dirs(os.path.join(args.out, "predictions")))
logger.info("make predictions")
data = is_tRNA(data, out_dir, args)
if args.coral:
logger.info("make CoRaL predictions")
run_coral(data, out_dir, args)
write_data(data[0], out_file)
logger.info("Done")
def cluster(args):
"""
Creating clusters
"""
args = _check_args(args)
read_stats_file = op.join(args.dir_out, "read_stats.tsv")
if file_exists(read_stats_file):
os.remove(read_stats_file)
bam_file, seq_obj = _clean_alignment(args)
logger.info("Parsing matrix file")
seqL, y, l = parse_ma_file(seq_obj, args.ffile)
# y, l = _total_counts(seqL.keys(), seqL)
logger.info("counts after: %s" % sum(y.values()))
logger.info("# sequences after: %s" % l)
dt = pd.DataFrame({'sample': y.keys(), 'counts': y.values()})
dt['step'] = 'aligned'
dt.to_csv(read_stats_file, sep="\t", index=False, header=False, mode='a')
if len(seqL.keys()) < 10:
logger.error(
"It seems you have low coverage. Please check your fastq files have enough sequences."
)
raise ValueError("So few sequences.")
logger.info("Cleaning bam file")
y, l = _total_counts(list(seqL.keys()), seqL)
logger.info("counts after: %s" % sum(y.values()))
logger.info("# sequences after: %s" % l)
dt = pd.DataFrame({'sample': y.keys(), 'counts': y.values()})
dt['step'] = 'cleaned'
dt.to_csv(read_stats_file, sep="\t", index=False, header=False, mode='a')
clusL = _create_clusters(seqL, bam_file, args)
y, l = _total_counts(list(clusL.seq.keys()), clusL.seq, aligned=True)
logger.info("counts after: %s" % sum(y.values()))
logger.info("# sequences after: %s" % l)
dt = pd.DataFrame({'sample': y.keys(), 'counts': y.values()})
dt['step'] = 'clusters'
dt.to_csv(read_stats_file, sep="\t", index=False, header=False, mode='a')
logger.info("Solving multi-mapping events in the network of clusters")
clusLred = _cleaning(clusL, args.dir_out)
y, l = _total_counts(clusLred.clus, seqL)
logger.info("counts after: %s" % sum(y.values()))
logger.info("# sequences after: %s" % l)
dt = pd.DataFrame({'sample': y.keys(), 'counts': y.values()})
dt['step'] = 'meta-cluster'
dt.to_csv(read_stats_file, sep="\t", index=False, header=False, mode='a')
logger.info("Clusters up to %s" % (len(clusLred.clus.keys())))
if args.show:
logger.info("Creating sequences alignment to precursor")
clusLred = show_seq(clusLred, args.index)
clusLred = peak_calling(clusLred)
clusLred = _annotate(args, clusLred)
logger.info("Creating json and count matrix")
json_file = _create_json(clusLred, args)
logger.info("Output file in: %s" % args.dir_out)
if args.db:
name = args.db + ".db"
logger.info("Create database: database/" + name)
data = load_data(json_file)
out_dir = op.join(args.dir_out, "database")
make_database(data, name, out_dir)
logger.info("Finished")
def cluster(args):
"""
Creating clusters
"""
args = _check_args(args)
read_stats_file = op.join(args.dir_out, "read_stats.tsv")
if file_exists(read_stats_file):
os.remove(read_stats_file)
bam_file, seq_obj = _clean_alignment(args)
logger.info("Parsing matrix file")
seqL, y, l = parse_ma_file(seq_obj, args.ffile)
# y, l = _total_counts(seqL.keys(), seqL)
logger.info("counts after: %s" % sum(y.values()))
logger.info("# sequences after: %s" % l)
dt = pd.DataFrame({'sample': y.keys(), 'counts': y.values()})
dt['step'] = 'aligned'
dt.to_csv(read_stats_file, sep="\t", index=False, header=False, mode='a')
if len(seqL.keys()) < 10:
logger.error("It seems you have low coverage. Please check your fastq files have enough sequences.")
raise ValueError("So few sequences.")
logger.info("Cleaning bam file")
y, l = _total_counts(list(seqL.keys()), seqL)
logger.info("counts after: %s" % sum(y.values()))
logger.info("# sequences after: %s" % l)
dt = pd.DataFrame({'sample': y.keys(), 'counts': y.values()})
dt['step'] = 'cleaned'
dt.to_csv(read_stats_file, sep="\t", index=False, header=False, mode='a')
clusL = _create_clusters(seqL, bam_file, args)
y, l = _total_counts(list(clusL.seq.keys()), clusL.seq, aligned=True)
logger.info("counts after: %s" % sum(y.values()))
logger.info("# sequences after: %s" % l)
dt = pd.DataFrame({'sample': y.keys(), 'counts': y.values()})
dt['step'] = 'clusters'
dt.to_csv(read_stats_file, sep="\t", index=False, header=False, mode='a')
logger.info("Solving multi-mapping events in the network of clusters")
clusLred = _cleaning(clusL, args.dir_out)
y, l = _total_counts(clusLred.clus, seqL)
logger.info("counts after: %s" % sum(y.values()))
logger.info("# sequences after: %s" % l)
dt = pd.DataFrame({'sample': y.keys(), 'counts': y.values()})
dt['step'] = 'meta-cluster'
dt.to_csv(read_stats_file, sep="\t", index=False, header=False, mode='a')
logger.info("Clusters up to %s" % (len(clusLred.clus.keys())))
if args.show:
logger.info("Creating sequences alignment to precursor")
clusLred = show_seq(clusLred, args.index)
clusLred = peak_calling(clusLred)
clusLred = _annotate(args, clusLred)
logger.info("Creating json and count matrix")
json_file = _create_json(clusLred, args)
logger.info("Output file in: %s" % args.dir_out)
if args.db:
name = args.db + ".db"
logger.info("Create database: database/" + name)
data = load_data(json_file)
out_dir = op.join(args.dir_out, "database")
make_database(data, name, out_dir)
logger.info("Finished")
