def get_concidered_exons(cnx, exon_class, regulation):
"""
Get the exon considered
:param cnx: (sqlite3 connect object) connection to sed databse
:param exon_class: (string) None, GC, AT or GC-AT, the exons that we consider, if \
exon_class is set to None, all exon regulated by a splicing factor are considered
:param regulation: (string) up or down
:return: (list of string/list 2 int) list of sf_name if exon_class is None or list of exons if \
exon_class is not None.
"""
if exon_class is None:
list_sf = group_factor.get_wanted_sf_name(None)
return list_sf
else:
if exon_class == "AT":
exon_list = difference(cnx, group_factor.at_rich_down, group_factor.gc_rich_down, regulation)
elif exon_class == "GC":
exon_list = difference(cnx, group_factor.gc_rich_down, group_factor.at_rich_down, regulation)
elif exon_class == "GC-AT":
exon_list = difference(cnx, group_factor.at_rich_down, group_factor.gc_rich_down, regulation)
exon_list += difference(cnx, group_factor.gc_rich_down, group_factor.at_rich_down, regulation)
print("exons GC-AT concidered : %s" % len(exon_list))
else:
exon_list = get_control_exon(cnx, exon_class)
return exon_list
def main(union, columns, name, sf_type, contrast, operation):
"""
Launch the main function.
"""
target_columns = columns.split(",")
exon_type = "CCE"
seddb = "/".join(os.path.realpath(__file__).split("/")[:-2]) + "/data/sed.db"
cnx = functions.connexion(seddb)
ctrl_dic, ctrl_full = control_exon_adapter.control_handler(cnx, exon_type, operation)
print("test")
if union != "union":
output = "/".join(os.path.realpath(__file__).split("/")[:-2]) + "/result/new_heatmap/"
# If the output directory does not exist, then we create it !
if not os.path.isdir(output):
os.mkdir(output)
id_projects, name_projects = group_factor.get_id_and_name_project_wanted(cnx, sf_type)
print(id_projects)
print(name_projects)
for regulations in [["down"]]:
# Creating heatmap
if sf_type is not None:
redundant_ag_at_and_u1_u2(cnx, regulations[0])
projects_tab, df_stat, project_names, new_targets = \
create_matrix(cnx, id_projects, name_projects, target_columns,
ctrl_dic, ctrl_full, regulations, operation, None, sf_type)
if len(new_targets) > 1:
heatmap_creator(np.array(projects_tab), new_targets, project_names, output, contrast, name)
else:
simple_heatmap(np.array(projects_tab), new_targets, project_names, output, contrast, name)
heatmap_gc_sorted(np.array(projects_tab), df_stat,
new_targets, project_names, output, contrast, name=name)
else:
output = "/".join(os.path.realpath(__file__).split("/")[:-2]) + "/result/new_heatmap_union/"
# If the output directory does not exist, then we create it !
if not os.path.isdir(output):
os.mkdir(output)
name_projects = group_factor.get_wanted_sf_name(sf_type)
for regulations in [["down"]]:
# Creating heatmap
if sf_type is not None:
redundant_ag_at_and_u1_u2(cnx, regulations[0])
projects_tab, df_stat, project_names, new_targets = \
create_matrix(cnx, None, name_projects, target_columns,
ctrl_dic, ctrl_full, regulations, operation, "union", sf_type)
if len(new_targets) > 1:
heatmap_creator(np.array(projects_tab), new_targets, project_names, output, contrast, name)
else:
simple_heatmap(np.array(projects_tab), new_targets, project_names, output, contrast, name)
heatmap_gc_sorted(np.array(projects_tab), df_stat, new_targets,
project_names, output, contrast, name)
def get_exon_regulated_by_sf(cnx, regulation):
"""
Get the exons ``regulation`` regulated by a splicing factors.
:param cnx: (sqlite3 connect object) connection to sed database
:param regulation: (string) up or down
:return: (list of list of 2 int) list of exons regulated by a splicing factor
"""
name_projects = group_factor.get_wanted_sf_name("all")
exon_list = []
for sf_name in name_projects:
exon_list += get_every_events_4_a_sl(cnx, sf_name, regulation)
exon_list = np.unique(exon_list, axis=0).tolist()
exon_list = [list(map(int, exon)) for exon in exon_list]
print("Number of exons regulated by a splicing factor : %s" %
len(exon_list))
return exon_list
def main(target_columns, nt_list, output_folder, seddb, exon_type,
size_bp_up_seq, regulation, name_tab):
if seddb is None:
seddb = os.path.realpath(os.path.dirname(__file__)).replace(
"src", "data/sed.db")
cnx = sqlite3.connect(seddb)
ctrl_dic, ctrl_full = control_exon_adapter.control_handler(
cnx, exon_type, size_bp_up_seq)
sf_names = group_factor.get_wanted_sf_name(cnx)
target_columns_new = [
target_columns[i].replace("iupac", "%s_nt" % nt_list[i])
for i in range(len(target_columns))
]
df = get_values_for_many_projects_iupac_dnt(cnx, sf_names,
target_columns_new, regulation,
ctrl_full, exon_type)
if output_folder:
df.to_csv("%s/%s" % (output_folder, name_tab), sep="\t", index=False)
return df
def main_2bc(target_columns, name, seddb, exon_type,
output, sf_type, sf_type2remove, regulation="down",
contrast=20, operation="mean", morder='C', mascending=True):
"""
:param target_columns: (list of str) list of columns of interest
:param name: (str) partial name of the figure
:param seddb: (str) path to sed database
:param exon_type: (str) the type of control exons we want
:param contrast: (int) the scale of contrast we want
:param operation: (str) the type of graphics we want (mean or median)
:param output: (str) the result folder
:param sf_type: (str) the type of splicing factor of interest
:param sf_type2remove: (str) the list of splicing factor we want to remove
:param regulation: (str) "down"
"""
global nt_list
nt_list = ["A", "T", "G", "C"]
global write_order
write_order = "N"
global ascending
ascending = mascending
global order
order = morder
cnx = functions.connexion(seddb)
ctrl_dic, ctrl_full = \
control_exon_adapter.control_handler(cnx, exon_type, operation)
name_projects = group_factor.get_wanted_sf_name(sf_type)
# Creating heatmap
projects_tab, df_stat, project_names, new_targets = \
create_matrix_bis(cnx, name_projects, target_columns,
ctrl_dic, ctrl_full, regulation, operation,
sf_type2remove)
heatmap_gc_sorted(np.array(projects_tab), df_stat, new_targets,
project_names, output, contrast, name)
def main():
"""
Launch the creation of figures.
"""
global exon_type
exon_type = "CCE"
seddb = "/".join(os.path.realpath(__file__).split("/")[:-2]) + "/data/sed.db"
fasterdb = "/".join(os.path.realpath(__file__).split("/")[:-2]) + "/data/fasterDB_lite.db"
regs = ["down"]
cnx = connexion(seddb)
cnx_fasterdb = connexion(fasterdb)
# columns = ["iupac_exon", "exon_size", "upstream_intron_size", "downstream_intron_size", "gene_size",
# "median_flanking_intron_size", "force_donor", "force_acceptor", "iupac_upstream_intron_adjacent1",
# "nb_intron_gene", "nb_good_bp_%s" % size_bp_up_seq, "hbound", "ag_count", "mfe_3ss", "mfe_5ss",
# "iupac_upstream_intron_ppt_area"]
columns = ["exon_size", "force_donor", "force_acceptor", "iupac_exon", "iupac_upstream_intron", "iupac_downstream_intron"]
ctrl_dic, ctrl_full = control_exon_adapter.control_handler(cnx, exon_type, size_bp_up_seq)
if len(sys.argv) < 2:
output = "/".join(os.path.realpath(__file__).split("/")[:-2]) + "/result/project_figures_new/"
# If the output directory does not exist, then we create it !
if not os.path.isdir(output):
os.mkdir(output)
id_projects, name_projects = get_interest_project(cnx)
for regulation in regs:
print(regulation)
for target_column in columns:
if target_column == "nb_good_bp":
if not os.path.isdir(output_bp):
os.mkdir(output_bp)
print(" %s" % target_column)
if "iupac" in target_column:
for nt in nt_dic.keys():
create_figure_iupac_dnt(cnx, id_projects, name_projects, target_column, regulation, output,
nt, ctrl_dic, ctrl_full)
elif "dnt" in target_column:
for dnt in dnt_dic.keys():
create_figure_iupac_dnt(cnx, id_projects, name_projects, target_column, regulation,
output, dnt, ctrl_dic, ctrl_full)
else:
create_figure(cnx, cnx_fasterdb, id_projects, name_projects, target_column, regulation, output,
ctrl_dic, output_bp, ctrl_full)
elif sys.argv[1] == "union":
output = "/".join(os.path.realpath(__file__).split("/")[:-2]) + "/result/project_figures_union_new/"
# If the output directory does not exist, then we create it !
if not os.path.isdir(output):
os.mkdir(output)
name_projects = group_factor.get_wanted_sf_name(cnx)
for regulation in regs:
print(regulation)
for target_column in columns:
if target_column == "nb_good_bp":
if not os.path.isdir(output_bp):
os.mkdir(output_bp)
print(" %s" % target_column)
if "iupac" in target_column:
for nt in nt_dic.keys():
create_figure_iupac_dnt(cnx, None, name_projects, target_column, regulation, output, nt,
ctrl_dic, ctrl_full,
"union")
elif "dnt" in target_column:
for dnt in dnt_dic.keys():
create_figure_iupac_dnt(cnx, None, name_projects, target_column, regulation, output, dnt,
ctrl_dic, ctrl_full,
"union")
else:
create_figure(cnx, cnx_fasterdb, None, name_projects, target_column, regulation, output, ctrl_dic,
output_bp, ctrl_full, "union")
else:
print("wrong arg !")
cnx.close()
cnx_fasterdb.close()
def main(level, xaxis, yaxis, name_fig, exon_type, nt_list, exon_class, operation, representation):
"""
Create the correlation matrix (gene_size vs iupac)
"""
seddb = "/".join(os.path.realpath(__file__).split("/")[:-2]) + "/data/sed.db"
cnx = functions.connexion(seddb)
ctrl_dic, full_dic = control_exon_adapter.control_handler(cnx, exon_type, operation)
nt_list = nt_list.split(",")
regulation = "down"
output = "/".join(os.path.realpath(__file__).split("/")[:-2]) + "/result/new_correlation/"
couple_targets = define_couple_targets(xaxis, yaxis, nt_list)
if level == "project":
name_fig += "_project"
id_projects, name_projects = group_factor.get_id_and_name_project_wanted(cnx, None)
# If the output directory does not exist, then we create it !
if not os.path.isdir(output):
os.mkdir(output)
for i in range(len(couple_targets)):
values_xaxis = get_axis_value(couple_targets[i][0], cnx, id_projects, ctrl_dic, regulation, operation,
representation)
values_yaxis = get_axis_value(couple_targets[i][1], cnx, id_projects, ctrl_dic, regulation, operation,
representation)
if len(values_xaxis) != len(values_yaxis):
print("Warning the list of value don't have the same length")
figure_creator(values_xaxis, values_yaxis, name_projects, regulation, couple_targets[i][0],
couple_targets[i][1], exon_type, output,
name_fig, representation, ctrl_dic)
elif level == "exon":
if exon_class is None:
list_sf = group_factor.get_wanted_sf_name(None)
else:
exon_list = get_concidered_exons(cnx, exon_class, regulation)
name_fig += "_%s_exons" % exon_class
name_fig += "_Exon_LVL"
if not os.path.isdir(output):
os.mkdir(output)
gene_columns = ["median_intron_size", "iupac_gene", "gene_size"]
if xaxis in gene_columns and yaxis in gene_columns:
for i in range(len(couple_targets)):
if exon_class is None:
value_xaxis, value_yaxis, exon_name = get_gene_values(cnx, list_sf, couple_targets[i][0],
couple_targets[i][1], regulation)
else:
value_xaxis, value_yaxis, exon_name = get_gene_values(cnx, exon_list, couple_targets[i][0],
couple_targets[i][1], regulation)
value_xaxis, value_yaxis, exon_name = remove_none_values(value_xaxis, value_yaxis, exon_name)
value_xaxis, name_x = get_relative_values(value_xaxis, ctrl_dic, couple_targets[i][0], representation)
value_yaxis, name_y = get_relative_values(value_yaxis, ctrl_dic, couple_targets[i][1], representation)
figure_creator_exon(value_xaxis, value_yaxis, regulation, name_x, name_y, exon_type,
exon_name, output, name_fig, representation)
density_creator_exon(value_xaxis, value_yaxis, regulation, name_x, name_y, exon_type, output, name_fig,
representation)
else:
for i in range(len(couple_targets)):
if exon_class is None:
value_xaxis, value_yaxis, exon_name = get_exons_values(cnx, list_sf, couple_targets[i][0],
couple_targets[i][1], regulation)
else:
value_xaxis, value_yaxis, exon_name = get_exons_values(cnx, exon_list, couple_targets[i][0],
couple_targets[i][1], regulation)
value_xaxis, value_yaxis, exon_name = remove_none_values(value_xaxis, value_yaxis, exon_name)
value_xaxis, value_yaxis, exon_name = remove_wrong_size_values(value_xaxis, couple_targets[i][0],
exon_name, value_yaxis,
couple_targets[i][1])
if len(value_xaxis) != len(value_yaxis) or len(value_xaxis) != len(exon_name):
print("Warning the list of value do'nt have the same length")
value_xaxis, name_x = get_relative_values(value_xaxis, ctrl_dic, couple_targets[i][0], representation)
value_yaxis, name_y = get_relative_values(value_yaxis, ctrl_dic, couple_targets[i][1], representation)
figure_creator_exon(value_xaxis, value_yaxis, regulation, name_x, name_y, exon_type,
exon_name, output, name_fig, representation)
density_creator_exon(value_xaxis, value_yaxis, regulation, name_x, name_y, exon_type, output, name_fig,
representation)
else:
name_fig += "_union"
list_sf = group_factor.get_wanted_sf_name(None)
if not os.path.isdir(output):
os.mkdir(output)
for i in range(len(couple_targets)):
values_xaxis = get_axis_value(couple_targets[i][0], cnx, list_sf, ctrl_dic, regulation, operation,
representation)
values_yaxis = get_axis_value(couple_targets[i][1], cnx, list_sf, ctrl_dic, regulation, operation,
representation)
if len(values_xaxis) != len(values_yaxis):
print("Warning the list of value don't have the same length")
figure_creator(values_xaxis, values_yaxis, list_sf, regulation, couple_targets[i][0],
couple_targets[i][1], exon_type, output,
name_fig, representation, ctrl_dic)