def create_othergc_exon_file(cnx, output):
"""
Create the GC-exons file regulated by SRSF2 HNRNPC and SRSF3.
:param cnx: (sqlite3 object) allow connection to sed database
:param output: (str) path were the exon files will be created
"""
output = output + "/other_GC_exons/"
if not os.path.isdir(output):
os.mkdir(output)
gc_exon_all = get_exons_list(cnx, group_factor.gc_rich_down, "down")
print("GC exons : %s" % len(gc_exon_all))
at_exon_all = get_exons_list(cnx, group_factor.at_rich_down, "down")
print("Number of at exons all: %s" % len(at_exon_all))
other_gc_all = get_exons_list(cnx, group_factor.other, "down")
venn_diagram_creator([other_gc_all, gc_exon_all, at_exon_all],
["hnRNPC & SRSF3 & SRSF2", "GC-exons", "AT-exons"],
output)
other_gc_list = [get_exons_list(cnx, [fact], "down")
for fact in group_factor.other]
full_list = other_gc_list + [gc_exon_all, at_exon_all]
full_names = list(group_factor.other) + ["GC_exons", "AT_exons"]
common = get_multi_list_common_exons(full_list)
for exon_list, list_name in zip(full_list, full_names):
start = len(exon_list)
print("%s : %s exons" % (list_name, start))
file_writer(exon_list, list_name + "_all", output)
exon_list = [exon for exon in exon_list if exon not in common]
stop = len(exon_list)
loss = round((start - stop) / start * 100, 1)
print("%s : %s exons after filter (loss of %s percent)" %
(list_name, stop, loss))
file_writer(exon_list, list_name, output)
def create_unregulated_exon_list(cnx, output, exon_type):
"""
Create the list of GC/At rich unregulated exons
:param cnx: (sqlite3 connect objecy) connection to sed database
:param output: (str) path were the exon files will be created
:param exon_type: (str) the type of control exons
"""
gc_exon_all = get_exons_list(cnx, group_factor.gc_rich_down, "down")
at_exon_all = get_exons_list(cnx, group_factor.at_rich_down, "down")
gc_exon = [exon for exon in gc_exon_all if exon not in at_exon_all]
at_exon = [exon for exon in at_exon_all if exon not in gc_exon_all]
first_exons = find_first_exons(cnx, exon_type)
last_exons = find_last_exons(cnx, exon_type)
exon2remove = gc_exon + at_exon + last_exons
print("exon to remove : %s" % len(exon2remove))
min_intron_size, gc_content = \
get_control_exon_information(cnx, exon_type, exon2remove)
gc_exon = get_exons_of_interest(cnx, exon_type, exon2remove,
min_intron_size, gc_content, "GC")
at_exon = get_exons_of_interest(cnx, exon_type, exon2remove,
min_intron_size, gc_content, "AT")
gc_exon = [exon for exon in gc_exon if exon not in at_exon]
at_exon = [exon for exon in at_exon if exon not in gc_exon]
print("gc_exon : %s exons" % len(gc_exon))
print("at_exon : %s exons" % len(at_exon))
file_writer(gc_exon, "GC_unregulated", output)
file_writer(at_exon, "AT_unregulated", output)
cnx.close()
def main():
"""
Make the enrichment analysis comparing the frequencies of exon regulated by splicesome factors \
for an AT and GC exons list.
"""
nb_iteration = 10
seddb = os.path.realpath(os.path.dirname(__file__)).replace(
"src/GC_AT_group_regulated_U1_U2", "data/sed.db")
cnx = sqlite3.connect(seddb)
output = os.path.realpath(os.path.dirname(__file__)).replace(
"src/GC_AT_group_regulated_U1_U2",
"result/GC_AT_group_regulated_U1_U2/")
if not os.path.isdir(output):
os.mkdir(output)
div_group = {
"AT_rich": group_factor.at_rich_down,
"GC_rich": group_factor.gc_rich_down,
"SNRPC": ["SNRPC"],
"SNRNP70": ["SNRNP70"],
"DDX5_17": ["DDX5_DDX17"],
"SF1": ["SF1"],
"U2AF1": ["U2AF1"],
"U2AF2": ['U2AF2'],
"SF3A3": ["SF3A3"],
"SF3B4": ["SF3B4"]
}
dic_exon = {}
for name_group in div_group.keys():
print("Getting all exon regulated by %s factor" % name_group)
dic_exon[name_group] = get_exons_list(cnx, div_group[name_group],
"down")
at_gc_intersection = exon_intersection(dic_exon["AT_rich"],
dic_exon["GC_rich"])
u1_u2_intersection = exon_intersection(
get_exons_list(cnx, group_factor.u1_factors, "down"),
get_exons_list(cnx, group_factor.u2_factors, "down"))
print("GC-AT group intersection : %s exons" % len(at_gc_intersection))
print("U1-U2 interesection : %s exons" % len(u1_u2_intersection))
dic_exon["GC_pure"] = exon_difference(dic_exon["GC_rich"],
at_gc_intersection)
dic_exon["AT_pure"] = exon_difference(dic_exon["AT_rich"],
at_gc_intersection)
for key in dic_exon:
print("%s : %s" % (key, len(dic_exon[key])))
dic_spliceosome = {}
for key in dic_exon.keys():
if "AT" not in key and "GC" not in key:
dic_spliceosome[key] = dic_exon[key]
analysis_dic, super_dict = analysis_maker(dic_exon["GC_pure"],
dic_exon["AT_pure"],
dic_spliceosome,
u1_u2_intersection, nb_iteration)
file_writer(output, analysis_dic, nb_iteration, "GC", "AT")
with open("%ssuper_dict_%s.txt" % (output, nb_iteration), "w") as outfile:
outfile.write("super_dict=%s\n" % str(super_dict))
cnx.close()
def create_ct_ga_rich_exon_list(cnx, output):
"""
Create the GA and the CT rich exons list.
:param cnx: (sqlite3 object) allow connection to sed database
:param output: (str) path were the exon files will be created
"""
ga_exon_all = get_exons_list(cnx, group_factor.ga_rich_down, "down")
ct_exon_all = get_exons_list(cnx, group_factor.ct_rich_down, "down")
ga_exon = [exon for exon in ga_exon_all if exon not in ct_exon_all]
ct_exon = [exon for exon in ct_exon_all if exon not in ga_exon_all]
print("ga_exon_all : %s exons" % len(ga_exon_all))
print("ga_exon : %s exons" % len(ga_exon))
print("ct_exon_all : %s exons" % len(ct_exon_all))
print("ct_exon : %s exons" % len(ct_exon))
file_writer(ga_exon, "GA_rich", output)
file_writer(ct_exon, "CT_rich", output)
def main():
"""
Create a bed file containing info about GC frequency of every GC-AT exons.
"""
base = os.path.dirname(
os.path.dirname(os.path.dirname(os.path.realpath(__file__))))
seddb = base + "/data/sed.db"
fasterdb = base + "/data/fasterDB_lite.db"
output = base + "/result/correlation_GC-AT-exons_TAD"
if not os.path.isdir(output):
os.mkdir(output)
cnx = sqlite3.connect(seddb)
u1_exons = [
list(map(int, exon)) for exon in get_exons_list(
cnx, ["SNRPC", "SNRNP70", "DDX5_DDX17"], "down")
]
u2_exons = [
list(map(int, exon)) for exon in get_exons_list(
cnx, ["U2AF2", "SF1", "SF3A3", "SF3B4"], "down")
]
print("U1-exons : %s exons" % len(u1_exons))
print("U2-exons : %s exons" % len(u2_exons))
exon_list = udf.get_exon_regulated_by_sf(cnx, "down")
print("Getting exon data ...")
exon_data = get_exon_info(cnx, seddb, fasterdb, exon_list, u1_exons,
u2_exons)
print("Writing bed")
write_bed(output, exon_data, "data_for_regulated_exons")
cnx.close()
cnx = sqlite3.connect(seddb)
exon2remove = udf.get_exon_regulated_by_sf(cnx, "down")
cnx_fasterdb = sqlite3.connect(fasterdb)
exon_list = get_control_exon_information(cnx_fasterdb, "CCE",
exon2remove) + exon2remove
cnx_fasterdb.close()
print("CCE exons + regulated exons : %s" % len(exon_list))
print("Getting CCE + regulated exon data ...")
exon_data = get_exon_info(cnx, seddb, fasterdb, exon_list, u1_exons,
u2_exons)
print("Writing bed")
write_bed(output, exon_data, "data_for_regulated_CCE_exons")
cnx.close()
def main_3g(list_file, name_file, seddb, output, reverse):
"""
Make the enrichment analysis comparing the frequencies of exon regulated by splicesome factors \
for an AT and GC exons list.
"""
nb_iteration = 10000
cnx = sqlite3.connect(seddb)
div_group = {
"SNRPC": ["SNRPC"],
"SNRNP70": ["SNRNP70"],
# "DDX5_17": ["DDX5_DDX17"],
"SF1": ["SF1"],
"U2AF1": ["U2AF1"],
"U2AF2": ['U2AF2'],
# "SF3A3": ["SF3A3"], "SF3B4": ["SF3B4"]
}
dic_exon = {}
for name_group in div_group.keys():
print("Getting all exon regulated by %s factor" % name_group)
dic_exon[name_group] = get_exons_list(cnx, div_group[name_group],
"down")
for key in dic_exon:
print("%s : %s" % (key, len(dic_exon[key])))
dic_spliceosome = {}
for key in dic_exon.keys():
if "AT" not in key and "GC" not in key:
dic_spliceosome[key] = dic_exon[key]
analysis_dic, super_dict = analysis_maker_bis(list_file, name_file,
dic_spliceosome,
nb_iteration)
filename = file_writer(output, analysis_dic, nb_iteration, name_file[0],
name_file[1])
fig_3g(filename, output, reverse)
with open("%ssuper_dict_%s.txt" % (output, nb_iteration), "w") as outfile:
outfile.write("super_dict=%s\n" % str(super_dict))
cnx.close()