diff --git a/src/nt_composition/make_nt_correlation.py b/src/nt_composition/make_nt_correlation.py
index c94e1080c24841e0bb743b5a79e022395a8f11cb..aae5b2122d1f439783d3bb82e4a33d92791a24c5 100644
--- a/src/nt_composition/make_nt_correlation.py
+++ b/src/nt_composition/make_nt_correlation.py
@@ -80,7 +80,9 @@ def get_project_colocalisation(cnx: sqlite3.Connection, project: str,
:param exon_bc: exons in big communities
:param inter_chr: True to only get inter-chromosomal interactions \
False else
- :param the feature that we want to recover
+ :param level: the feature that we want to recover
+ :param cell: Interactions are only selected from projects made on \
+ a specific cell line (ALL to disable this filter). (default ALL)
:return: The table containing the number of interaction by projects
"""
logging.debug(f'Recovering interaction ({os.getpid()})')
@@ -333,7 +335,7 @@ def create_density_fig(df: pd.DataFrame, project: str, ft_type: str, ft: str,
plt.xlabel(f"Freq of {ft} in {ml}")
plt.ylabel(f"Freq of {ft} in co-localized {level}s")
title = f'Freq of {ft} of {level}s and their co-localized partner in ' \
- f'{project}'
+ f'{project}'
if inter_chr:
title += '\n(inter chromosomal interactions)'
plt.title(title)
@@ -347,7 +349,7 @@ def create_density_fig(df: pd.DataFrame, project: str, ft_type: str, ft: str,
def create_density_figure(nt: str, ft_type: str,
project: str, weight: int, global_weight: int,
- same_gene: bool, inflation: float,
+ same_gene: bool, inflation: float, cell_line: str,
compute_mean: bool,
community_size: Optional[int],
inter_chr: bool = False, level: str = "exon",
@@ -368,6 +370,8 @@ def create_density_figure(nt: str, ft_type: str,
:param same_gene: Say if we consider as co-localised exon within the \
same gene
:param inflation: The inflation parameter
+ :param cell_line: Interactions are only selected from projects made on \
+ a specific cell line (ALL to disable this filter). (default ALL)
:param compute_mean: True to compute the mean frequency of co-localised \
exons, false to only compute the frequency of one co-localized exons.
:param community_size: The size of the community to consider in the \
@@ -386,7 +390,7 @@ def create_density_figure(nt: str, ft_type: str,
if not outfile.is_file():
exons_bc = recover_exon_in_big_communities(community_size, project,
weight, global_weight,
- inflation, level)
+ inflation, cell_line, level)
cnx = sqlite3.connect(ConfigNt.db_file)
arr_interaction = get_project_colocalisation(cnx, project, weight,
global_weight, same_gene,
@@ -453,15 +457,16 @@ def create_scatterplot(df_cor: pd.DataFrame, ft_type: str, ft: str,
plt.close()
-def execute_density_figure_function(di: pd.DataFrame, project : str,
+def execute_density_figure_function(di: pd.DataFrame, project: str,
ft_type: str, ft: str, weight: int,
global_weight: int,
same_gene: bool,
inflation: float,
+ cell_line: str,
compute_mean: bool,
community_size: Optional[int],
inter_chr: bool = False,
- level: str= "exon",
+ level: str = "exon",
) -> Dict[str, Any]:
"""
Execute create_density_figure and organized the results in a dictionary.
@@ -478,6 +483,8 @@ def execute_density_figure_function(di: pd.DataFrame, project : str,
:param same_gene: Say if we consider as co-localised exon within the \
same gene
:param inflation: The inflation parameter
+ :param cell_line: Interactions are only selected from projects made on \
+ a specific cell line (ALL to disable this filter). (default ALL)
:param compute_mean: True to compute the mean frequency of co-localised \
exons, false to only compute the frequency of one co-localized exons.
:param community_size: he size of the community to consider in the \
@@ -490,15 +497,15 @@ def execute_density_figure_function(di: pd.DataFrame, project : str,
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, inflation,
- compute_mean,
+ cell_line, compute_mean,
community_size, inter_chr, level)
if global_weight == 0:
return {"project": project, "ft_type": ft_type,
- "ft": ft, "cor": r, "pval": p,
- 'nb_interaction': di[di['projects'] == project].iloc[0, 1]}
+ "ft": ft, "cor": r, "pval": p,
+ 'nb_interaction': di[di['projects'] == project].iloc[0, 1]}
else:
return {"project": project, "ft_type": ft_type,
- "ft": ft, "cor": r, "pval": p}
+ "ft": ft, "cor": r, "pval": p}
def combine_dic(list_dic: List[Dict]) -> Dict:
@@ -518,6 +525,7 @@ def combine_dic(list_dic: List[Dict]) -> Dict:
def recover_exon_in_big_communities(community_size: Optional[int],
project: str, weight: int,
global_weight: int, inflation: float,
+ cell_line: str,
level: str) -> Optional[np.array]:
"""
Recover the list of exon present in community with a larger size than \
@@ -532,13 +540,15 @@ def recover_exon_in_big_communities(community_size: Optional[int],
seen in `global_weight` project are taken into account
:param weight: The weight of interaction to consider
:parma inflation: The inflation parameter
+ :param cell_line: Interactions are only selected from projects made on \
+ a specific cell line (ALL to disable this filter). (default ALL)
:param level: The kind of feature to analyse (exon or gene)
:return: The list of exon of interest
"""
if community_size is None:
return None
outfile = ConfigNt.get_community_file(project, weight, global_weight,
- True, inflation, level,
+ True, inflation, cell_line, level,
"_communities.txt")
if not outfile.is_file():
msg = f"The file {outfile} was not found ! You must try " \
@@ -555,7 +565,8 @@ def recover_exon_in_big_communities(community_size: Optional[int],
def create_all_frequency_figures(ps: int, weight: int = 1,
global_weight: int = 0, ft_type: str = "nt",
- same_gene = True, inflation: float = 1.5,
+ same_gene=True, inflation: float = 1.5,
+ cell_line: str = "ALL",
compute_mean: bool = True,
community_size: Optional[int] = None,
inter_chr: bool = False, level: str = "exon",
@@ -573,6 +584,8 @@ def create_all_frequency_figures(ps: int, weight: int = 1,
:param same_gene: Say if we consider as co-localised exon within the \
same gene
:param inflation: The inflation parameter
+ :param cell_line: Interactions are only selected from projects made on \
+ a specific cell line (ALL to disable this filter). (default ALL)
:param compute_mean: True to compute the mean frequency of co-localised \
exons, false to only compute the frequency of one co-localized exons.
:param community_size: The size of the community to consider in the \
@@ -596,7 +609,8 @@ 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,
- inflation, compute_mean, community_size, inter_chr, level]
+ inflation, cell_line, compute_mean, community_size, inter_chr,
+ level]
processes.append(pool.apply_async(execute_density_figure_function,
args))
results = [proc.get(timeout=None) for proc in processes]