def redundant_ag_at_and_u1_u2(cnx, regulation):
"""
Create the list of redundant exons between the AT and GC rich list of exons and \
between the U1 and U2 list of exons
:param cnx: (sqlite3 connect object) allow connection to sed database
:param regulation: (string) the regulation we want for the common exons
:return: (list of list of 2 int) list of exons identified by their gene id and their exons position
"""
exon_at = []
for sf_name in group_factor.at_rich_down:
exon_at += union_dataset_function.get_every_events_4_a_sl(cnx, sf_name, regulation)
exon_at = union_dataset_function.washing_events_all(exon_at)
exon_gc = []
for sf_name in group_factor.gc_rich_down:
exon_gc += union_dataset_function.get_every_events_4_a_sl(cnx, sf_name, regulation)
exon_gc = union_dataset_function.washing_events_all(exon_gc)
global redundant_gc_at
redundant_gc_at = [exon for exon in exon_at if exon in exon_gc]
print("redundant exon GC and AT rich : %s" % len(redundant_gc_at))
exon_u1 = []
for sf_name in group_factor.u1_factors:
exon_u1 += union_dataset_function.get_every_events_4_a_sl(cnx, sf_name, regulation)
exon_u1 = union_dataset_function.washing_events_all(exon_u1)
exon_u2 = []
for sf_name in group_factor.u2_factors:
exon_u2 += union_dataset_function.get_every_events_4_a_sl(cnx, sf_name, regulation)
exon_u2 = union_dataset_function.washing_events_all(exon_u2)
global redundant_u1_u2
redundant_u1_u2 = [exon for exon in exon_u1 if exon in exon_u2]
print("redundant exon U1 and U2 rich : %s" % len(redundant_u1_u2))
def get_exon_list(cnx, annotation_name, regulation):
"""
Get the exon_list wanted.
:param cnx: (sqlite3 connect object) connection to sed database
:param annotation_name: (string) GC-AT or a sf_name
:param regulation: (string) the regulation of an exon list by a factor(s)
:return: (list of 2 int) gene id and exon_pos
"""
if "GC" in annotation_name or "AT" in annotation_name:
annotation_name = annotation_name.split("_")[0]
folder = os.path.realpath(os.path.dirname(__file__)).replace(
"src", "data/")
my_file = "%s%s_rich_exons" % (folder, annotation_name)
exon_list = extract_exon_list(my_file)
elif "U1-FACTORS" in annotation_name or "U2-FACTORS" in annotation_name:
annotation_name = annotation_name.split("_")[0]
dic_name = {
"U1-FACTORS": ["SNRPC", "SNRNP70", "DDX5_DDX17"],
"U2-FACTORS": ["U2AF2", "SF1", "SF3A3", "SF3B4"]
}
exon_list = []
for sf_name in dic_name[annotation_name]:
exon_list += union_dataset_function.get_every_events_4_a_sl(
cnx, sf_name, regulation)
exon_list = union_dataset_function.washing_events_all(exon_list)
else:
annotation_name = annotation_name.split("_")[0]
sf_name = annotation_name.upper()
sf_name = sf_name.replace("SFRS",
"SRSF").replace("TRA2A", "TRA2A_B").replace(
"DDX5-17", "DDX5_DDX17")
exon_list = union_dataset_function.get_every_events_4_a_sl(
cnx, sf_name, regulation)
return exon_list
def extract_data(cnx, cnx_sed, list_files, list_names, pos, regulation):
"""
:param cnx: (sqlite3 connect object) connection to fasterDB lite
:param cnx_sed: (sqlite3 connect object) connection to sed
:param list_files: (list of string) list of files containing exon set
:param list_names: (list of string) the name of exon set
:param pos: (int) the position of interest within the list ``list_files`` and ``list_names``. \
Those 2 lists must have the same length
:param regulation: (string) up or down
:return: (list of ExonClass object) list of exon.
"""
if list_files:
exon_list = extract_exon_files(cnx, list_files[pos])
else:
dic_name = {
"U1-factors": ["SNRPC", "SNRNP70", "DDX5_DDX17"],
"U2-factors": ["U2AF2", "SF1", "SF3A3", "SF3B4"]
}
exon_list_tmp = []
for sf_name in dic_name[list_names[pos]]:
exon_list_tmp += union_dataset_function.get_every_events_4_a_sl(
cnx_sed, sf_name, regulation)
exon_list_tmp = union_dataset_function.washing_events_all(
exon_list_tmp)
exon_list = [
exon_class_bp.ExonClass(cnx, str(exon[0]), int(exon[0]),
int(exon[1])) for exon in exon_list_tmp
]
print("%s : %s %s exons" % (list_names[pos], len(exon_list), regulation))
return exon_list
def difference(cnx, list1, list2, regulation):
"""
Return the exons regulated by the factors in list1 if they are not regulated by the factors in list2
:param cnx: (sqlite3 connect object) connection to sed database
:param list1: (list of string) list of splicing factors
:param list2: (list of strings) list of splicing factros
:param regulation: (string) the exons with the regulation ``regulation`` regulated by the splicing factors in \
``list1`` or ``list2``
:return:(list of list of 2 int
"""
exon_list1 = []
exon_list2 = []
for sf_name in list1:
exon_list1 += union_dataset_function.get_every_events_4_a_sl(cnx, sf_name, regulation)
exon_list1 = union_dataset_function.washing_events_all(exon_list1)
for sf_name in list2:
exon_list2 += union_dataset_function.get_every_events_4_a_sl(cnx, sf_name, regulation)
exon_list2 = union_dataset_function.washing_events_all(exon_list2)
return [exon for exon in exon_list1 if exon not in exon_list2]
def get_gene_values(cnx, sf_list, target_column1, target_column2, regulation):
"""
Return the values of target_column in every`\
`regulation`` exons regulated by a splicing factor in (one or multiple) cell lines.
:param cnx: (sqlite3 connexion object) allow connexion to sed database
:param sf_list: (list of string) the list of splicing factor studied
:param target_column1: (string) the value for which we want to get the median value for the ``regulation`` \
exon.
:param target_column2: (string) the value for which we want to get the median value for the ``regulation`` \
exon.
:param regulation: (list of string) up or down or up + down
:return: 3 lists :
* values : (list of list of float) each sublist corresponds to the value of `` target_column`` for \
every exons regulated by a splicing factor
* exon_name : (list of list of string) each sublist corresponds to the name of \
every exons regulated by a splicing factor - the value in the sublist **i** position **j** \
in the ``value`` and ``exon_name`` corresponds to the same exons
* all_sf (list of string) list of each sf studied
"""
if "$" in target_column1:
target_column1, nt1 = target_column1.split("$")
else:
nt1 = None
if "$" in target_column2:
target_column2, nt2 = target_column2.split("$")
else:
nt2 = None
exon_list = []
if isinstance(sf_list[0], str):
for sf_name in sf_list:
exon_list += union_dataset_function.get_every_events_4_a_sl(cnx, sf_name, regulation)
exon_list = union_dataset_function.washing_events_all(exon_list)
else:
exon_list = sf_list
gene_id = []
for val in exon_list:
if val[0] not in gene_id:
gene_id.append(val[0])
gene_name = [union_dataset_function.get_gene_name(cnx, my_id) for my_id in gene_id]
if nt1:
values1 = get_list_of_value_iupac_dnt(cnx, exon_list, target_column1, nt1)
else:
values1 = functions.get_list_of_value(cnx, exon_list, target_column1)
if nt2:
values2 = get_list_of_value_iupac_dnt(cnx, exon_list, target_column2, nt2)
else:
values2 = functions.get_list_of_value(cnx, exon_list, target_column2)
return values1, values2, gene_name
def get_exons_list(cnx, sf_list, regulation):
"""
Return every non-redundant exons regulated by at least one factor in ``sf_list`` (with the regulation \
``regulation``)
:param cnx: (sqlite3 connexion object) allow connexion to sed database
:param sf_list: (list of string) the list of splicing factor studied
:param regulation: (list of string) up or down or up + down
:return: (list of list of int) list of exons shownig the regulation ``regulation`` at least for a factor \
in ``sf_list``
"""
exon_list = []
for sf_name in sf_list:
exon_list += union_dataset_function.get_every_events_4_a_sl(
cnx, sf_name, regulation)
exon_list = union_dataset_function.washing_events_all(exon_list)
return exon_list
def get_exons_values(cnx, sf_list, target_column1, target_column2, regulation):
"""
Return the values of target_column in every` \
`regulation`` exons regulated by a splicing factor in (one or multiple) cell lines.
:param cnx: (sqlite3 connexion object) allow connexion to sed database
:param sf_list: (list of string) the list of splicing factor studied
:param target_column1: (string) the value for which we want to get the median value for the ``regulation`` exon.
:param target_column2: (string) the value for which we want to get the median value for the ``regulation`` exon.
:param regulation: (list of string) up or down or up + down
:return: 3 lists :
* values : (list of list of float) each sublist corresponds to the value of `` target_column`` for \
every exons regulated by a splicing factor
* exon_name : (list of list of string) each sublist corresponds to the name of \
every exons regulated by a splicing factor - the value in the sublist **i** position **j** \
in the ``value`` and ``exon_name`` corresponds to the same exons
* all_sf (list of string) list of each sf studied
"""
if "$" in target_column1:
target_column1, nt1 = target_column1.split("$")
else:
nt1 = None
if "$" in target_column2:
target_column2, nt2 = target_column2.split("$")
else:
nt2 = None
exon_list = []
if isinstance(sf_list[0], str):
for sf_name in sf_list:
exon_list += union_dataset_function.get_every_events_4_a_sl(cnx, sf_name, regulation)
exon_list = union_dataset_function.washing_events_all(exon_list)
else:
exon_list = sf_list
print(len(exon_list))
exon_name = ["%s_%s" % (union_dataset_function.get_gene_name(cnx, a[0]), a[1]) for a in exon_list]
values1 = get_interest_values(cnx, exon_list, target_column1, nt1)
values2 = get_interest_values(cnx, exon_list, target_column2, nt2)
if len(exon_name) * 2 == len(values1):
exon_name = ["%s_upstream" % a for a in exon_name] + \
["%s_downstream" % a for a in exon_name]
return values1, values2, exon_name