diff --git a/src/nt_composition/__main__.py b/src/nt_composition/__main__.py
index 72e99bbc3c22e8de467f9686b467009d397de89b..8948a419cae14906965339c7d97260551c1f709e 100644
--- a/src/nt_composition/__main__.py
+++ b/src/nt_composition/__main__.py
@@ -14,14 +14,14 @@ from .get_projects_interaction import get_interactions_number
import lazyparser as lp
-@lp.flag(same_gene=True)
+@lp.flag(same_gene=True, inter_chr=True)
@lp.parse(weight='weight > 0',
ft_type=['nt', 'dnt', 'tnt', 'codon', 'aa', 'properties'],
kind=["density", "distance"])
def launcher(weight: int = 1, global_weight: int = 0, ft_type: str = 'nt',
same_gene: bool = False, compute_mean: bool = True,
iteration: int = 10000, kind: str = 'density',
- community_size: int = -1,
+ community_size: int = -1, inter_chr: bool = False,
logging_level: str = "DISABLE"):
"""
Launch the creation of density file.
@@ -44,21 +44,23 @@ def launcher(weight: int = 1, global_weight: int = 0, ft_type: str = 'nt',
frequency of the same feature for other co-localized exons. 'distance' \
to check if the frequencies of co-localized exon are closer than \
randomly expected. (default : 'density').*
- :param community_size: consider only exons within communities bigger
+ :param community_size: consider only exons within communities bigger \
than community size. if community_size = -1. This option is deactivated.
+ :param inter_chr: True to only get inter-chromosomal interactions \
+ False else (default: False)
:param logging_level: The level of data to display (default 'DISABLE')
"""
- get_interactions_number(weight, same_gene, logging_level)
+ get_interactions_number(weight, same_gene, inter_chr, logging_level)
if community_size == -1:
community_size = None
if kind == "density":
create_all_frequency_figures(ConfigNt.cpu, weight, global_weight,
ft_type, same_gene, compute_mean,
- community_size,
+ community_size, inter_chr,
logging_level)
else:
create_all_distance_figures(ConfigNt.cpu, weight, global_weight,
- ft_type, same_gene, iteration,
+ ft_type, same_gene, iteration, inter_chr,
logging_level)
diff --git a/src/nt_composition/config.py b/src/nt_composition/config.py
index 8dc88e86ae8954ff51d879f6f0bbc039e10c7f61..a7a222d43a53683b4c2e29579ce1644fcc86cdb8 100644
--- a/src/nt_composition/config.py
+++ b/src/nt_composition/config.py
@@ -14,7 +14,7 @@ from ..find_interaction_cluster.config import ConfigGraph
def get_weight_folder(weight: int, global_weight: int, ft_type: str,
- community_size: Optional[int]):
+ community_size: Optional[int], inter_chr: bool):
"""
Get the weight folder.
@@ -25,20 +25,23 @@ def get_weight_folder(weight: int, global_weight: int, ft_type: str,
seen in `global_weight` project are taken into account
:parma ft_type: The type fo featrue of interst
:param community_size: The minimum size of communities we want to consider
+ :param inter_chr: True to only get inter-chromosomal interactions \
+ False else
:return: The folder that will contains the interaction with a weight \
greater or equal to `weigh` in ChIA-PET projects
"""
+ inter_str = "_inter-chromosmal" if inter_chr else ""
if community_size is None:
c = ''
else:
c = f'_community_size_gte_{community_size}'
if global_weight == 0:
weight_folder = ConfigNt.density_folder / \
- f"{ft_type}_project_weight-{weight}{c}"
+ f"{ft_type}_project_weight-{weight}{c}{inter_str}"
else:
weight_folder = ConfigNt.density_folder / \
f"{ft_type}_weight-{weight}_" \
- f"global_weight-{global_weight}{c}"
+ f"global_weight-{global_weight}{c}{inter_str}"
weight_folder.mkdir(parents=True, exist_ok=True)
return weight_folder
@@ -46,7 +49,8 @@ def get_weight_folder(weight: int, global_weight: int, ft_type: str,
def get_density_file(weight: int, global_weight: int, project: str,
ft_type: str, ft: str, fig: bool = False,
kind: str = 'density',
- community_size: Optional[int] = None):
+ community_size: Optional[int] = None,
+ inter_chr: bool = False):
"""
Get the filename that will contain the density data or figure.
@@ -61,6 +65,8 @@ def get_density_file(weight: int, global_weight: int, project: str,
:param fig: Say if the result file is a figure or not
:param kind: The kind of the figure
:param community_size: The minimum size of communities we want to consider
+ :param inter_chr: True to only get inter-chromosomal interactions \
+ False else
:return: The filename that will contain the density data or figure.
"""
if fig:
@@ -68,14 +74,15 @@ def get_density_file(weight: int, global_weight: int, project: str,
else:
ext = "txt"
res_folder = get_weight_folder(weight, global_weight, ft_type,
- community_size) / project
+ community_size, inter_chr) / project
res_folder.mkdir(exist_ok=True, parents=True)
return res_folder / f"{project}_{ft_type}_{ft}_{kind}.{ext}"
def get_density_recap(weight: int, global_weight: int,
ft_type: str, ft: str, fig: bool = False,
- community_size: Optional[int] = False):
+ community_size: Optional[int] = False,
+ inter_chr: bool = False):
"""
Get the density correlation recap file.
@@ -88,6 +95,8 @@ def get_density_recap(weight: int, global_weight: int,
:param ft: A feature
:param fig: Say if the result file is a figure or not
:param community_size: The minimum size of communities we want to consider
+ :param inter_chr: True to only get inter-chromosomal interactions \
+ False else
:return:
"""
if fig:
@@ -95,22 +104,29 @@ def get_density_recap(weight: int, global_weight: int,
else:
ext = "txt"
outfolder = get_weight_folder(weight, global_weight, ft_type,
- community_size)
+ community_size, inter_chr)
outfolder.mkdir(exist_ok=True, parents=True)
return outfolder / f"{ft_type}_{ft}_density_recap.{ext}"
-def get_interaction_file(weight: int):
+def get_interaction_file(weight: int, same_gene: bool, inter_chr: bool):
"""
Return the interaction file : coresponding to the file containing \
iteraction with a weight greater or eaqual to `weight`
:param weight: minimum weight of interaction to \
concider
+ :param same_gene: Say if we are considering interaction within the same \
+ gene
+ :param inter_chr: True to only get inter-chromosomal interactions \
+ False else
:return: The interaction filename
"""
+ same_str = "_same_gene_allowed" if same_gene else ""
+ inter_str = "inter-chromosmal_" if inter_chr else ""
return ConfigNt.interaction / \
- f"interaction_number_weight>={weight}_by_project.txt"
+ f"{inter_str}interaction_number_weight>={weight}_by_project" \
+ f"{same_str}.txt"
def get_features(ft_type: str) -> List[str]:
diff --git a/src/nt_composition/distance.py b/src/nt_composition/distance.py
index 83aa3fcf67bd91f9919d18637affd3cbd8b148b5..5ec3ebdbc1b0695e8f98cfcd80c5a3ed9a9b5e40 100644
--- a/src/nt_composition/distance.py
+++ b/src/nt_composition/distance.py
@@ -254,7 +254,7 @@ def create_distance_figure(df: pd.DataFrame, project: str, figfile: Path,
def create_figures(ft: str, ft_type: str,
project: str, weight: int, global_weight: int,
- same_gene: bool, iteration: int,
+ same_gene: bool, iteration: int, inter_chr: bool = False,
logging_level: str = "DISABLE"
) -> None:
"""
@@ -271,24 +271,29 @@ def create_figures(ft: str, ft_type: str,
seen in `global_weight` project are taken into account
:param same_gene: Say if we consider as co-localised exon within the \
same gene
+ :param inter_chr: True to only get inter-chromosomal interactions \
+ False else
:param logging_level: The level of information to display
:param iteration: The number of iteration
"""
logging_def(ConfigNt.interaction, __file__, logging_level)
outfile = ConfigNt.get_density_file(weight, global_weight, project,
ft_type, ft, fig=False,
- kind="distance")
+ kind="distance",
+ inter_chr=inter_chr)
if not outfile.is_file():
cnx = sqlite3.connect(ConfigNt.db_file)
arr_interaction = get_project_colocalisation(cnx, project, weight,
- global_weight, same_gene)
+ global_weight, same_gene,
+ inter_chr=inter_chr)
dic_freq = get_frequency_dic(cnx, ft, ft_type)
df = create_distance_table(arr_interaction, dic_freq, project,
iteration, ft, ft_type)
df.to_csv(outfile, sep="\t", index=False)
figfile = ConfigNt.get_density_file(weight, global_weight, project,
ft_type, ft, fig=True,
- kind="distance")
+ kind="distance",
+ inter_chr=inter_chr)
create_distance_figure(df, project, figfile, ft, iteration)
@@ -296,7 +301,8 @@ def execute_density_figure_function(project: str,
ft_type: str, ft: str, weight: int,
global_weight: int,
same_gene: bool,
- iteration: int) -> None:
+ iteration: int,
+ inter_chr: bool = False) -> None:
"""
Execute create_density_figure and organized the results in a dictionary.
@@ -312,16 +318,19 @@ def execute_density_figure_function(project: str,
same gene
:param iteration: The number of iteration of interest to create \
the control set.
+ :param inter_chr: True to only get inter-chromosomal interactions \
+ False else
"""
logging.info(f'Working on {project}, {ft_type}, {ft} - {os.getpid()}')
create_figures(ft, ft_type, project, weight,
- global_weight, same_gene, iteration)
+ global_weight, same_gene, iteration, inter_chr)
def create_all_distance_figures(ps: int, weight: int = 1,
global_weight: int = 0, ft_type: str = "nt",
same_gene: bool = True,
iteration: int = 10000,
+ inter_chr: bool = False,
logging_level: str = "DISABLE"):
"""
Make density figure for every selected projects.
@@ -336,6 +345,8 @@ def create_all_distance_figures(ps: int, weight: int = 1,
:param same_gene: Say if we consider as co-localised exon within the \
same gene
:param iteration: Number of iteration for control exons
+ :param inter_chr: True to only get inter-chromosomal interactions \
+ False else
:param ps: The number of processes to create
"""
logging_def(ConfigNt.interaction, __file__, logging_level)
@@ -349,7 +360,7 @@ def create_all_distance_figures(ps: int, weight: int = 1,
processes = []
for project, ft, ft_type in param:
args = [project, ft_type, ft, weight, global_weight, same_gene,
- iteration]
+ iteration, inter_chr]
processes.append(pool.apply_async(execute_density_figure_function,
args))
for proc in processes:
diff --git a/src/nt_composition/get_projects_interaction.py b/src/nt_composition/get_projects_interaction.py
index f9dbc9dfedfad312bd320f1141fe9669af4535ba..002446d8c3137635b733e587b3d2fa6104d86092 100644
--- a/src/nt_composition/get_projects_interaction.py
+++ b/src/nt_composition/get_projects_interaction.py
@@ -22,7 +22,8 @@ import logging
def get_interaction_by_project(cnx: sqlite3.Connection, weight: int,
- same_gene: bool) -> pd.DataFrame:
+ same_gene: bool,
+ inter_chr: bool = False) -> pd.DataFrame:
"""
Get the number of interactions by projects.
@@ -30,13 +31,18 @@ def get_interaction_by_project(cnx: sqlite3.Connection, weight: int,
:param weight: A weight threshold
:param same_gene: Say if we are considering interaction within the same \
gene
+ :param inter_chr: True to only get inter-chromosomal interactions \
+ False else
:return: The table containing the number of interaction by projects
"""
logging.debug('Getting interaction from database')
+ inter_str = " AND level = 'inter'" if inter_chr else ""
+ inter_str_t = " AND t1.level = 'inter'" if inter_chr else ""
if same_gene:
query = f"SELECT id_project, COUNT(*) " \
f"FROM cin_exon_interaction " \
f"WHERE weight >= {weight} " \
+ f"{inter_str} " \
f"GROUP BY id_project"
else:
query = f"""SELECT id_project, COUNT(*)
@@ -45,6 +51,7 @@ def get_interaction_by_project(cnx: sqlite3.Connection, weight: int,
AND t1.exon1 = t2.id
AND t1.exon2 = t3.id
AND t2.id_gene != t3.id_gene
+ {inter_str_t}
GROUP BY id_project"""
df = pd.read_sql_query(query, cnx)
df.columns = ['projects', 'interaction_count']
@@ -53,13 +60,18 @@ def get_interaction_by_project(cnx: sqlite3.Connection, weight: int,
return df
-def make_barplot(df: pd.DataFrame, weight: int):
+def make_barplot(df: pd.DataFrame, weight: int,
+ same_gene: bool, inter_chr: bool):
"""
Make a barplot displaying the number of interactions for every project.
:param weight: The minimum weight of interaction to concidere them
:param df: The dataframe containing the number of interaction by \
projects
+ :param same_gene: Say if we are considering interaction within the same \
+ gene
+ :param inter_chr: True to only get inter-chromosomal interactions \
+ False else
"""
logging.debug("Creating barplot figure")
ConfigNt.interaction.mkdir(parents=True, exist_ok=True)
@@ -68,7 +80,7 @@ def make_barplot(df: pd.DataFrame, weight: int):
plt.figure(figsize=(20, 12))
sns.barplot(x="projects", y="interaction_count", data=df)
plt.xticks(rotation=90)
- outfile = ConfigNt.get_interaction_file(weight)
+ outfile = ConfigNt.get_interaction_file(weight, same_gene, inter_chr)
plt.savefig(outfile.parent / outfile.name.replace('txt', 'pdf'),
bbox_inches='tight')
plt.close()
@@ -91,6 +103,7 @@ def select_projects(df: pd.DataFrame):
def get_interactions_number(weight: int = 1, same_gene: bool = False,
+ inter_chr: bool = False,
logging_level: str = "DISABLE"):
"""
Get the number of interaction by projects
@@ -98,13 +111,15 @@ def get_interactions_number(weight: int = 1, same_gene: bool = False,
:param weight: The minimum weight of correlation to consider them
:param same_gene: Say if we are considering interaction within the same \
gene
+ :param inter_chr: True to only get inter-chromosomal interactions \
+ False else
"""
logging_def(ConfigNt.interaction, __file__, logging_level)
logging.info(f'Recovering interaction count with a weight of {weight}')
cnx = sqlite3.connect(ConfigNt.db_file)
- df = get_interaction_by_project(cnx, weight, same_gene)
- make_barplot(df, weight)
- df.to_csv(ConfigNt.get_interaction_file(weight),
+ df = get_interaction_by_project(cnx, weight, same_gene, inter_chr)
+ make_barplot(df, weight, same_gene, inter_chr)
+ df.to_csv(ConfigNt.get_interaction_file(weight, same_gene, inter_chr),
sep="\t", index=False)
sns.barplot()
select_projects(df)
diff --git a/src/nt_composition/make_nt_correlation.py b/src/nt_composition/make_nt_correlation.py
index ef7ceb8b372371c9b417a52cdf35d3f115b045b0..2c94056cbc7644dbf17fe1ae6c1d66f8b700813b 100644
--- a/src/nt_composition/make_nt_correlation.py
+++ b/src/nt_composition/make_nt_correlation.py
@@ -47,11 +47,11 @@ def get_select_addition(global_weight: int, get_weight: bool, same_gene: bool):
if get_weight:
if global_weight == 0 and same_gene:
return ', weight'
- elif global_weight == 0 and not same_gene:
+ elif global_weight == 0:
return ', t1.weight'
elif global_weight >= 0 and same_gene:
return ', ROUND(AVG(weight), 2)'
- elif global_weight >= 0 and not same_gene:
+ elif global_weight >= 0:
return ', ROUND(AVG(t1.weight), 2)'
return ''
@@ -60,7 +60,8 @@ def get_project_colocalisation(cnx: sqlite3.Connection, project: str,
weight: int,
global_weight: int, same_gene: bool,
get_weight: bool = False,
- exon_bc: Optional[np.array] = None) -> np.array:
+ exon_bc: Optional[np.array] = None,
+ inter_chr: bool = False) -> np.array:
"""
Get the interactions in project `project`
@@ -75,13 +76,16 @@ def get_project_colocalisation(cnx: sqlite3.Connection, project: str,
:param same_gene: Say if we consider as co-localised exon within the \
same gene
:param exon_bc: exons in big communities
+ :param inter_chr: True to only get inter-chromosomal interactions \
+ False else
:return: The table containing the number of interaction by projects
"""
logging.debug(f'Recovering interaction ({os.getpid()})')
select_add = get_select_addition(global_weight, get_weight, same_gene)
+ inter_str = " AND level = 'inter'" if inter_chr else ""
+ inter_str_t = " AND t1.level = 'inter'" if inter_chr else ""
if exon_bc is None:
- filter_exons = ''
- filter_exons_t = filter_exons
+ filter_exons = filter_exons_t = ''
else:
tmp = tuple(exon_bc)
filter_exons = f' AND exon1 IN {tmp} AND exon2 IN {tmp}'
@@ -91,7 +95,9 @@ def get_project_colocalisation(cnx: sqlite3.Connection, project: str,
query = f"SELECT exon1, exon2{select_add} " \
"FROM cin_exon_interaction " \
f"WHERE weight >= {weight} " \
- f"AND id_project = '{project}'" + filter_exons
+ f"AND id_project = '{project}'" \
+ + filter_exons \
+ + inter_str
else:
query = f"""SELECT t1.exon1, t1.exon2{select_add}
FROM cin_exon_interaction t1, cin_exon t2, cin_exon t3
@@ -99,15 +105,18 @@ def get_project_colocalisation(cnx: sqlite3.Connection, project: str,
AND id_project = '{project}'
AND t1.exon1 = t2.id
AND t1.exon2 = t3.id
- AND t2.id_gene != t3.id_gene""" + filter_exons_t
+ AND t2.id_gene != t3.id_gene""" \
+ + filter_exons_t \
+ + inter_str_t
else:
good_projects = tuple(ConfigNt.good_projects)
if same_gene:
query = f"SELECT exon1, exon2{select_add} " \
f"FROM cin_exon_interaction " \
f"WHERE weight >= {weight} " \
- f"AND id_project IN {good_projects}" \
- f"{filter_exons}" \
+ f"AND id_project IN {good_projects} " \
+ f"{filter_exons} " \
+ f"{inter_str} " \
f"GROUP BY exon1, exon2 " \
f"HAVING COUNT(*) >= {global_weight}"
else:
@@ -119,6 +128,7 @@ def get_project_colocalisation(cnx: sqlite3.Connection, project: str,
AND t2.id_gene != t3.id_gene
AND t1.id_project IN {good_projects}
{filter_exons_t}
+ {inter_str_t}
GROUP BY exon1, exon2
HAVING COUNT(*) >= {global_weight}"""
@@ -258,7 +268,8 @@ def create_density_table(arr_interaction: np.array, dic_freq: Dict[str, float],
def create_density_fig(df: pd.DataFrame, project: str, ft_type: str, ft: str,
weight: int, global_weight: int,
- community_size: Optional[int]) -> Tuple[float, float]:
+ community_size: Optional[int],
+ inter_chr: bool) -> Tuple[float, float]:
"""
Compute a density file showing if the feature frequency of \
an exons correlates with the frequency in other co-localised exons.
@@ -274,6 +285,8 @@ def create_density_fig(df: pd.DataFrame, project: str, ft_type: str, ft: str,
the global weight is equal to 0 then then density figure are calculated \
by project, else all projet are merge together and the interaction \
seen in `global_weight` project are taken into account
+ :param inter_chr: True to only get inter-chromosomal interactions \
+ False else
:return: The correlation and the p-value
"""
logging.debug('Creating the density figure')
@@ -290,11 +303,15 @@ def create_density_fig(df: pd.DataFrame, project: str, ft_type: str, ft: str,
plt.legend()
plt.xlabel(f"Freq of {ft} in an exon")
plt.ylabel(f"Freq of {ft} in co-localized exons")
- plt.title(f'Freq of {ft} of exons and their co-localized partner in '
- f'{project}')
+ title = f'Freq of {ft} of exons and their co-localized partner in ' \
+ f'{project}'
+ if inter_chr:
+ title += '\n(inter chromosomal interactions)'
+ plt.title(title)
plt.savefig(ConfigNt.get_density_file(weight, global_weight, project,
ft_type, ft, fig=True,
- community_size=community_size))
+ community_size=community_size,
+ inter_chr=inter_chr))
plt.close()
return r, p
@@ -303,6 +320,7 @@ def create_density_figure(nt: str, ft_type: str,
project: str, weight: int, global_weight: int,
same_gene: bool, compute_mean: bool,
community_size: Optional[int],
+ inter_chr: bool = False,
logging_level: str = "DISABLE"
) -> Tuple[float, float]:
"""
@@ -323,13 +341,16 @@ def create_density_figure(nt: str, ft_type: str,
exons, false to only compute the frequency of one co-localized exons.
:param community_size: The size of the community to consider in the \
analysis.
+ :param inter_chr: True to only get inter-chromosomal interactions \
+ False else
:param logging_level: The level of information to display
:return: The correlation and the p-value
"""
logging_def(ConfigNt.interaction, __file__, logging_level)
outfile = ConfigNt.get_density_file(weight, global_weight, project,
ft_type, nt, fig=False,
- community_size=community_size)
+ community_size=community_size,
+ inter_chr=inter_chr)
if not outfile.is_file():
exons_bc = recover_exon_in_big_communities(community_size, project,
weight,
@@ -337,12 +358,13 @@ def create_density_figure(nt: str, ft_type: str,
cnx = sqlite3.connect(ConfigNt.db_file)
arr_interaction = get_project_colocalisation(cnx, project, weight,
global_weight, same_gene,
- False, exons_bc)
+ False, exons_bc,
+ inter_chr)
dic_freq = get_frequency_dic(cnx, nt, ft_type)
df = create_density_table(arr_interaction, dic_freq, compute_mean)
df.to_csv(outfile, sep="\t", index=False)
r, p = create_density_fig(df, project, ft_type, nt, weight,
- global_weight, community_size)
+ global_weight, community_size, inter_chr)
else:
logging.debug(f'The file {outfile} exist, recovering data '
f'({os.getpid()})')
@@ -353,7 +375,8 @@ def create_density_figure(nt: str, ft_type: str,
def create_scatterplot(df_cor: pd.DataFrame, ft_type: str, ft: str,
weight: int, global_weight: int,
- community_size: Optional[int]):
+ community_size: Optional[int],
+ inter_chr: bool):
"""
Create a scatterplot showing the R-value according to the \
number of interaction.
@@ -366,6 +389,8 @@ def create_scatterplot(df_cor: pd.DataFrame, ft_type: str, ft: str,
the global weight is equal to 0 then then density figure are calculated \
by project, else all projet are merge together and the interaction \
seen in `global_weight` project are taken into account
+ :param inter_chr: True to only get inter-chromosomal interactions \
+ False else
"""
sns.set()
sns.set_context('talk')
@@ -381,11 +406,15 @@ def create_scatterplot(df_cor: pd.DataFrame, ft_type: str, ft: str,
df_cor.project.values[i], fontsize=8)
plt.xlabel(f"Correlation for {ft} ({ft_type}) in project")
plt.ylabel("Number of total interaction in projects")
- plt.title(f'Project correlation for {ft} ({ft_type}) '
- f'according to the number of interaction in projects')
+ title = f'Project correlation for {ft} ({ft_type}) ' \
+ f'according to the number of interaction in projects'
+ if inter_chr:
+ title += "\n(inter-chromosmal interactions)"
+ plt.title(title)
plt.savefig(ConfigNt.get_density_recap(weight, global_weight, ft_type,
ft, fig=True,
- community_size=community_size))
+ community_size=community_size,
+ inter_chr=inter_chr))
plt.close()
@@ -394,7 +423,9 @@ def execute_density_figure_function(di: pd.DataFrame, project : str,
global_weight: int,
same_gene: bool,
compute_mean: bool,
- community_size: Optional[int]) -> Dict[str, Any]:
+ community_size: Optional[int],
+ inter_chr: bool = False
+ ) -> Dict[str, Any]:
"""
Execute create_density_figure and organized the results in a dictionary.
@@ -413,12 +444,14 @@ def execute_density_figure_function(di: pd.DataFrame, project : str,
exons, false to only compute the frequency of one co-localized exons.
:param community_size: he size of the community to consider in the \
analysis.
+ :param inter_chr: True to only get inter-chromosomal interactions \
+ False else
:return:
"""
logging.info(f'Working on {project}, {ft_type}, {ft} - {os.getpid()}')
r, p = create_density_figure(ft, ft_type, project, weight,
global_weight, same_gene, compute_mean,
- community_size)
+ community_size, inter_chr)
if global_weight == 0:
tmp = {"project": project, "ft_type": ft_type,
"ft": ft, "cor": r, "pval": p,
@@ -482,6 +515,7 @@ def create_all_frequency_figures(ps: int, weight: int = 1,
global_weight: int = 0, ft_type: str = "nt",
same_gene = True, compute_mean: bool = True,
community_size: Optional[int] = None,
+ inter_chr: bool = False,
logging_level: str = "DISABLE"):
"""
Make density figure for every selected projects.
@@ -500,10 +534,13 @@ def create_all_frequency_figures(ps: int, weight: int = 1,
:param community_size: The size of the community to consider in the \
analysis.
:param ps: The number of processes to create
+ :param inter_chr: True to only get inter-chromosomal interactions \
+ False else
"""
logging_def(ConfigNt.interaction, __file__, logging_level)
if global_weight == 0:
- di = pd.read_csv(ConfigNt.get_interaction_file(weight), sep="\t")
+ di = pd.read_csv(ConfigNt.get_interaction_file(weight, same_gene,
+ inter_chr), sep="\t")
projects = ConfigNt.good_projects
else:
di = pd.DataFrame()
@@ -514,22 +551,21 @@ def create_all_frequency_figures(ps: int, weight: int = 1,
processes = []
for project, ft, ft_type in param:
args = [di, project, ft_type, ft, weight, global_weight, same_gene,
- compute_mean, community_size]
+ compute_mean, community_size, inter_chr]
processes.append(pool.apply_async(execute_density_figure_function,
args))
- results = []
- for proc in processes:
- results.append(proc.get(timeout=None))
+ results = [proc.get(timeout=None) for proc in processes]
dic = combine_dic(results)
df_corr = pd.DataFrame(dic)
df_corr.to_csv(ConfigNt.get_density_recap(weight, global_weight, ft_type,
'ALL', fig=False,
- community_size=community_size),
+ community_size=community_size,
+ inter_chr=inter_chr),
sep="\t")
if global_weight == 0:
for ft in ft_list:
create_scatterplot(df_corr, ft_type, ft, weight, global_weight,
- community_size)
+ community_size, inter_chr)
if __name__ == "__main__":