diff --git a/src/find_interaction_cluster/__main__.py b/src/find_interaction_cluster/__main__.py
index 89ebea84808a1910969d84ec5a0c1057c538a258..f43a18e9d31c8b13631a7e0b78fdbbde93308d67 100644
--- a/src/find_interaction_cluster/__main__.py
+++ b/src/find_interaction_cluster/__main__.py
@@ -28,6 +28,7 @@ def launcher(weight: int = 1,
global_weight: int = 0,
same_gene: bool = True,
inflation: float = 1.5,
+ cell_line: str = 'ALL',
use_weight: bool = False,
ps: int = ConfigGraph.cpu,
html_fig: bool = False, feature: str = 'exon', region: str = '',
@@ -48,6 +49,8 @@ def launcher(weight: int = 1,
:param same_gene: Say if we consider as co-localised, exons within the \
same gene (True) or not (False) (default True)
: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 use_weight: Say if we want to write the weight into the result file.
:param html_fig: True to display the html figure (default False).
:param feature: The feature we want to analyse (default 'exon')
@@ -74,21 +77,21 @@ def launcher(weight: int = 1,
install_hipmcl("INFO")
multiple_community_launcher(weight, global_weight, project, same_gene,
- inflation, use_weight,
+ inflation, cell_line, use_weight,
html_fig, feature, logging_level)
# multiple_stat_launcher(ps, weight, global_weight, project, same_gene,
- # inflation, feature, logging_level)
+ # inflation, cell_line, feature, logging_level)
multiple_nt_lm_launcher(ps, weight, global_weight, project,
- same_gene, inflation, feature, region,
+ same_gene, inflation, cell_line, feature, region,
component_type, test_type, iteration, display_size,
logging_level=logging_level)
# if feature == "gene":
# # ppi_stat_launcher(weight, global_weight, project, same_gene,
- # # inflation,
+ # # inflation, cell_line,
# # ConfigGraph.ppi_threshold, iteration,
# # logging_level)
# coloc_ppi_stat_main(weight, global_weight, project, same_gene,
- # inflation, iteration, logging_level)
+ # inflation, cell_line, iteration, logging_level)
diff --git a/src/find_interaction_cluster/colocalisation_n_ppi_analysis.py b/src/find_interaction_cluster/colocalisation_n_ppi_analysis.py
index cd33bb836743019a844cc1b0f5c93d709fe6af72..18fe454254a7c9771f96ebf97007c0f316690c09 100644
--- a/src/find_interaction_cluster/colocalisation_n_ppi_analysis.py
+++ b/src/find_interaction_cluster/colocalisation_n_ppi_analysis.py
@@ -303,7 +303,7 @@ def create_figure(df_full: pd.DataFrame, outfile: Path) -> None:
def coloc_ppi_stat_main(weight: int, global_weight: int,
project: str, same_gene: bool, inflation: float,
- iteration: int = 1000,
+ cell_line: str = "ALL", iteration: int = 1000,
logging_level: str = "DISABLE"):
"""
Launch the statistical tests allowing to determine if interaction between \
@@ -318,6 +318,8 @@ def coloc_ppi_stat_main(weight: int, global_weight: int,
same gene
:param threshold: The minimum threshold needed to consider the interaction
:param iteration: The number of iteration to make
+ :param cell_line: Interactions are only selected from projects made on \
+ a specific cell line (ALL to disable this filter).
:param logging_level: Level of information to display
"""
ConfigGraph.community_folder.mkdir(exist_ok=True, parents=True)
@@ -329,13 +331,15 @@ def coloc_ppi_stat_main(weight: int, global_weight: int,
community_file = ConfigGraph.get_community_file(project, weight,
global_weight,
same_gene, inflation,
+ cell_line,
"gene", f".txt")
df_com = pd.read_csv(community_file, sep="\t")
df_com = df_com[df_com['nodes'] >= 10].copy()
full_df = create_scored_dataframe(df_com, ConfigPPI.fasterdb_ppi)
outfile = ConfigGraph.get_community_file(project, weight, global_weight,
- same_gene, inflation, "gene",
+ same_gene, inflation, cell_line,
+ "gene",
f"_interation_gene-protein.pdf",
"community_gene_vs_protein")
create_figure(full_df, outfile)
diff --git a/src/find_interaction_cluster/community_finder.py b/src/find_interaction_cluster/community_finder.py
index 9ff5afe0678c196e8658e72255ab03234bcd8296..c5c3a770bc93429fb5a762eff480a9971115b016 100644
--- a/src/find_interaction_cluster/community_finder.py
+++ b/src/find_interaction_cluster/community_finder.py
@@ -276,7 +276,8 @@ def get_figure_title(project, weight, global_weight, same_gene, feature):
def write_interaction_file(arr_interaction: np.array, project: str,
weight: int, global_weight: int, same_gene: bool,
- inflation: float, use_weight: bool = False,
+ inflation: float, cell_line: str = "ALL",
+ use_weight: bool = False,
feature: str = 'exon'):
"""
@@ -293,14 +294,18 @@ def write_interaction_file(arr_interaction: np.array, project: str,
:param use_weight: Say if we want to write the weight into the result file.
:param feature: Says if we want to work at gene or at exons level
: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).
:return:
"""
logging.debug('Writing interaction files ...')
outfile = ConfigGraph.get_community_file(project, weight, global_weight,
- same_gene, inflation, feature,
+ same_gene, inflation, cell_line,
+ feature,
f"_interation.txt")
result = ConfigGraph.get_community_file(project, weight, global_weight,
- same_gene, inflation, feature,
+ same_gene, inflation, cell_line,
+ feature,
f"_communities.txt")
with outfile.open('w') as f:
for exon1, exon2, cweight in arr_interaction:
@@ -313,7 +318,7 @@ def write_interaction_file(arr_interaction: np.array, project: str,
def community_finder(weight: int, global_weight: int, project: str = "",
same_gene=True, html_fig: bool = False,
feature: str = "exon", inflation: float = 1.5,
- use_weight: bool = True,
+ cell_line: str = "ALL", use_weight: bool = True,
logging_level: str = "DISABLE"):
"""
Find communities inside co-localisation between exons found in \
@@ -332,24 +337,28 @@ def community_finder(weight: int, global_weight: int, project: str = "",
:param inflation: The inflation parameter
:param use_weight: Say if we want to write the weight into the result file
:param feature: The feature we want to analyse (default 'exon')
+ :param cell_line: Interactions are only selected from projects made on \
+ a specific cell line (ALL to disable this filter).
"""
ConfigGraph.output_folder.mkdir(exist_ok=True, parents=True)
logging_def(ConfigGraph.output_folder, __file__, logging_level)
cnx = sqlite3.connect(ConfigGraph.db_file)
interaction = get_project_colocalisation(cnx, project, weight,
global_weight, same_gene, True,
- level=feature)
+ level=feature, cell=cell_line)
outfile, result_file = write_interaction_file(interaction, project,
weight, global_weight,
same_gene, feature=feature,
inflation=inflation,
- use_weight=use_weight)
+ use_weight=use_weight,
+ cell_line=cell_line)
graph = create_graph(interaction)
df, dic_community = find_communities(graph, project, outfile, result_file,
feature, inflation)
logging.debug('Writing results ...')
outfiles = [ConfigGraph.get_community_file(
- project, weight, global_weight, same_gene, inflation, feature, ext)
+ project, weight, global_weight, same_gene, inflation, cell_line,
+ feature, ext)
for ext in [f'.txt', f'.cyjs', f'.html']]
df.to_csv(outfiles[0], sep="\t", index=False)
logging.debug("Saving the graph ...")
@@ -383,6 +392,7 @@ def multiple_community_launcher(weight: int,
project: str,
same_gene: bool,
inflation: float = 1.5,
+ cell_line: str = "ALL",
use_weight: bool = False,
html_fig: bool = False,
feature: str = 'exon',
@@ -395,6 +405,8 @@ def multiple_community_launcher(weight: int,
: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).
:param use_weight: Say if we want to write the weight into the result file
:param html_fig: True to create an html figure, false else
:param feature: The feature we want to analyse (default 'exon')
@@ -407,4 +419,4 @@ def multiple_community_launcher(weight: int,
logging.info(f'Finding community for project : {project}, '
f'global_weight : {global_weight}, weight: {weight}')
community_finder(weight, global_weight, project, same_gene, html_fig,
- feature, inflation, use_weight)
+ feature, inflation, cell_line, use_weight)
diff --git a/src/find_interaction_cluster/config.py b/src/find_interaction_cluster/config.py
index 3e6c717cc65766ba2d4b90461d4ca98a104ed40b..7b72748eb19c408737845e9a9ee51544794cfb17 100644
--- a/src/find_interaction_cluster/config.py
+++ b/src/find_interaction_cluster/config.py
@@ -13,7 +13,8 @@ from typing import List, Dict
import pandas as pd
-def get_weight_folder(weight: int, global_weight: int, inflation: float):
+def get_weight_folder(weight: int, global_weight: int, inflation: float,
+ cell_line: str):
"""
Get the weight folder.
@@ -23,22 +24,26 @@ def get_weight_folder(weight: int, global_weight: int, inflation: float):
by project, else all project are merge together and the interaction \
seen in `global_weight` project are taken into account
: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).
:return: The folder that will contains the interaction with a weight \
greater or equal to `weigh` in ChIA-PET projects
"""
+ cell_line = "" if cell_line == "ALL" else f"-{cell_line}"
if global_weight == 0:
weight_folder = ConfigGraph.community_folder / \
- f"project_weight-{weight}-{inflation}"
+ f"project_weight-{weight}-{inflation}{cell_line}"
else:
weight_folder = ConfigGraph.community_folder / \
f"weight-{weight}_" \
- f"global_weight-{global_weight}-{inflation}"
+ f"global_weight-{global_weight}-{inflation}{cell_line}"
weight_folder.mkdir(parents=True, exist_ok=True)
return weight_folder
def get_community_file(project: str, weight: int, global_weight: int,
- same_gene: bool, inflation: float, feature: str = 'exon',
+ same_gene: bool, inflation: float,
+ cell_line: str = "ALL", feature: str = 'exon',
ext: str = ".txt", sub_fold: str = ''):
"""
Get the output file of interest.
@@ -52,12 +57,14 @@ def get_community_file(project: str, weight: int, global_weight: int,
: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).
:param feature: the kind of feature analyzed
:param ext: The file extension
:param sub_fold: if filled, then the data are recovered from a subfolder
:return: The filename of interest
"""
- folder = get_weight_folder(weight, global_weight, inflation)
+ folder = get_weight_folder(weight, global_weight, inflation, cell_line)
if sub_fold != '':
folder = folder / sub_fold
folder.mkdir(exist_ok=True, parents=True)
diff --git a/src/find_interaction_cluster/create_ppi_files.py b/src/find_interaction_cluster/create_ppi_files.py
index fa159e8984a8688bfdbc8aca4ae79d1621d49127..a990afa4978203cfca869e6484d86b80c958f09a 100644
--- a/src/find_interaction_cluster/create_ppi_files.py
+++ b/src/find_interaction_cluster/create_ppi_files.py
@@ -19,7 +19,7 @@ from .community_finder import create_graph, find_communities, \
write_cytoscape_graph
from doctest import testmod
from .radomization_test_ppi import get_ppi_community_gene, \
- get_dna_community_gene, update_overlap_df,summary_randomisation_test
+ get_dna_community_gene, update_overlap_df, summary_randomisation_test
def get_community_dic_fom_community_file(mfile: Union[Path, pd.DataFrame],
@@ -65,7 +65,8 @@ def ppi_array(fasterdb_ppi: Path, threshold: int) -> np.array:
def write_interaction_ppi(arr_interaction: np.array, project: str,
weight: int, global_weight: int, same_gene: bool,
- inflation: float, use_weight: bool = False):
+ inflation: float, cell_line: str = "ALL",
+ use_weight: bool = False):
"""
:param arr_interaction: Each couples of co-localized feature within a \
@@ -79,15 +80,19 @@ def write_interaction_ppi(arr_interaction: np.array, project: str,
:param same_gene: Say if we consider as co-localised, exons within the \
same gene (True) or not (False) (default False)
: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 use_weight: Say if we want to write the weight into the result file.
"""
logging.debug('Writing interaction files ...')
outfile = ConfigGraph.get_community_file(project, weight, global_weight,
- same_gene, inflation, "gene",
+ same_gene, inflation, cell_line,
+ "gene",
f"_interation_PPI_tmp.txt",
"community_gene_vs_protein")
result = ConfigGraph.get_community_file(project, weight, global_weight,
- same_gene, inflation, "gene",
+ same_gene, inflation, cell_line,
+ "gene",
f"_communities_PPI_tmp.txt",
"community_gene_vs_protein")
with outfile.open('w') as f:
@@ -100,7 +105,8 @@ def write_interaction_ppi(arr_interaction: np.array, project: str,
def ppi_community_finder(fasterdb_ppi: Path, project: str,
weight: int, global_weight: int,
- same_gene: bool = True, inflation: float=1.5,
+ same_gene: bool = True, inflation: float = 1.5,
+ cell_line: str = "ALL",
threshold: int = 700):
"""
Find communities inside protein-protein interaction file
@@ -115,19 +121,22 @@ def ppi_community_finder(fasterdb_ppi: Path, project: str,
:param same_gene: Say if we consider as co-localised, exons within the \
same gene (True) or not (False) (default False)
: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 threshold: The minimum threshold needed to consider the interaction
"""
interaction = ppi_array(fasterdb_ppi, threshold)
outfile, result_file = write_interaction_ppi(interaction, project,
weight, global_weight,
- same_gene, inflation)
+ same_gene, inflation,
+ cell_line)
graph = create_graph(interaction)
df, dic_community = find_communities(graph, project, outfile, result_file,
"gene")
logging.debug('Writing results ...')
outfiles = [ConfigGraph.get_community_file(
- project, weight, global_weight, same_gene, inflation, "gene", ext,
- "community_gene_vs_protein")
+ project, weight, global_weight, same_gene, inflation, cell_line,
+ "gene", ext, "community_gene_vs_protein")
for ext in [f'_graph_community_PPI.txt', f'_graph_community_PPI.cyjs']]
df.to_csv(outfiles[0], sep="\t", index=False)
logging.debug("Saving the graph ...")
@@ -229,7 +238,7 @@ def create_community_ppi_table(community_file: Path, fasterdb_ppi: Path,
project: str,
weight: int, global_weight: int,
same_gene: bool = True, inflation: float = 1.5,
- threshold: int = 700):
+ cell_line: str = "ALL", threshold: int = 700):
"""
:param community_file: A file containing community of gene interacting \
@@ -245,18 +254,21 @@ def create_community_ppi_table(community_file: Path, fasterdb_ppi: Path,
: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 threshold: The minimum threshold needed to consider the interaction
:return:
"""
outfile = ConfigGraph.get_community_file(project, weight, global_weight,
- same_gene, inflation, "gene",
+ same_gene, inflation, cell_line,
+ "gene",
'_graph_community_PPI.txt',
"community_gene_vs_protein")
if not outfile.is_file():
logging.debug("Creating the community file for PPI")
df_comm_ppi = ppi_community_finder(fasterdb_ppi, project, weight,
global_weight, same_gene, inflation,
- threshold)
+ cell_line, threshold)
else:
df_comm_ppi = pd.read_csv(outfile, sep="\t")
logging.debug("Turning ppi community file into a dic")
@@ -277,6 +289,7 @@ def ppi_stats_analysis(community_file: Path, fasterdb_ppi: Path,
project: str,
weight: int, global_weight: int,
same_gene: bool = True, inflation: float = 1.5,
+ cell_line: str = "ALL",
threshold: int = 700, iteration: int = 1000):
"""
:param community_file: A file containing community of gene interacting \
@@ -292,24 +305,28 @@ def ppi_stats_analysis(community_file: Path, fasterdb_ppi: Path,
: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 threshold: The minimum threshold needed to consider the interaction
:param iteration: The number of iteration to make
:return:
"""
outfile = ConfigGraph.get_community_file(project, weight,
global_weight, same_gene,
- inflation, "gene",
+ inflation, cell_line, "gene",
f"ppi_gene_complete_table.txt",
"community_gene_vs_protein")
ppi_outfile = ConfigGraph.get_community_file(project, weight,
global_weight,
- same_gene, inflation, "gene",
+ same_gene, inflation,
+ cell_line, "gene",
'_graph_community_PPI.txt',
"community_gene_vs_protein")
if not outfile.is_file():
df = create_community_ppi_table(community_file, fasterdb_ppi,
project, weight, global_weight,
- same_gene, inflation, threshold)
+ same_gene, inflation, cell_line,
+ threshold)
df.to_csv(outfile, sep="\t", index=False)
else:
df = pd.read_csv(outfile, sep="\t")
@@ -321,7 +338,8 @@ def ppi_stats_analysis(community_file: Path, fasterdb_ppi: Path,
df, dic_values = update_overlap_df(df_overlap, dic_dna_gene, ppi_gene,
iteration)
outstat = ConfigGraph.get_community_file(project, weight, global_weight,
- same_gene, inflation, "gene",
+ same_gene, inflation, cell_line,
+ "gene",
f"ppi_gene_table_{iteration}_"
f"stat.txt",
"community_gene_vs_protein")
@@ -330,7 +348,8 @@ def ppi_stats_analysis(community_file: Path, fasterdb_ppi: Path,
iteration, dic_values,
use_seed=False)
outstat = ConfigGraph.get_community_file(project, weight, global_weight,
- same_gene, inflation, "gene",
+ same_gene, inflation, cell_line,
+ "gene",
f"ppi_gene_table_{iteration}_"
f"stat_recap.txt",
"community_gene_vs_protein")
@@ -341,6 +360,7 @@ def ppi_stat_launcher(weight: int,
global_weight: int,
project: str,
same_gene: bool, inflation: float,
+ cell_line: str= "ALL",
threshold: int = 700,
iteration: int = 1000,
logging_level: str = "DISABLE"):
@@ -356,6 +376,8 @@ def ppi_stat_launcher(weight: int,
: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 threshold: The minimum threshold needed to consider the interaction
:param iteration: The number of iteration to make
:param logging_level: Level of information to display
@@ -369,10 +391,10 @@ def ppi_stat_launcher(weight: int,
community_file = ConfigGraph.get_community_file(project, weight,
global_weight,
same_gene, inflation,
- "gene", f".txt")
+ cell_line, "gene", f".txt")
ppi_stats_analysis(community_file, ConfigPPI.fasterdb_ppi, project,
- weight, global_weight, same_gene, inflation, threshold,
- iteration)
+ weight, global_weight, same_gene, inflation, cell_line,
+ threshold, iteration)
if __name__ == "__main__":
diff --git a/src/find_interaction_cluster/nt_and_community.py b/src/find_interaction_cluster/nt_and_community.py
index d1d591ed527c40e3dac73f386d52e1c17a2a3141..9aadb7e95052e95453f03e05b3cc146778a22323 100644
--- a/src/find_interaction_cluster/nt_and_community.py
+++ b/src/find_interaction_cluster/nt_and_community.py
@@ -249,8 +249,8 @@ def prepare_dataframe(df: pd.DataFrame, test_type: str, nt: str,
def create_outfiles(project: str, weight: int, global_weight: int,
- same_gene: bool, inflation: float, feature: str,
- cpnt_type: str, cpnt: str, test_type: str,
+ same_gene: bool, inflation: float, cell_line: str,
+ feature: str, cpnt_type: str, cpnt: str, test_type: str,
community_file: str) -> Tuple[Path, Path]:
"""
Create a file used to store diagnostics and a file used to store the \
@@ -269,6 +269,8 @@ def create_outfiles(project: str, weight: int, global_weight: int,
:param cpnt: The component (nt, aa, dnt) of interest
:param feature: The kind of feature analysed
: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).
:param test_type: The type of test to make (permutation or lm)
:param community_file: A file containing custom communities. If \
it equals to '' then weight, global weight and same genes parameter are \
@@ -288,12 +290,12 @@ def create_outfiles(project: str, weight: int, global_weight: int,
return outfile, outfile_ctrl
outfile = ConfigGraph.\
get_community_file(project, weight, global_weight, same_gene,
- inflation, feature,
+ inflation, cell_line, feature,
f"{cpnt}-{cpnt_type}_stat_{test_type}.txt",
outfolder)
outfile_ctrl = ConfigGraph.\
get_community_file(project, weight, global_weight, same_gene,
- inflation, feature,
+ inflation, cell_line, feature,
f"{cpnt}-{cpnt_type}_VS_CTRL_stat_{test_type}.pdf",
outfolder)
return outfile, outfile_ctrl
@@ -301,7 +303,7 @@ def create_outfiles(project: str, weight: int, global_weight: int,
def get_stat_cpnt_communities(df: pd.DataFrame, project: str, weight: int,
global_weight: int, same_gene: bool,
- inflation: float,
+ inflation: float, cell_line: str,
cpnt_type: str, cpnt: str,
dic_com: Dict, feature: str = "exon",
test_type: str = "",
@@ -323,6 +325,8 @@ def get_stat_cpnt_communities(df: pd.DataFrame, project: str, weight: int,
:param same_gene: Say if we consider as co-localised, exons within the \
same gene (True) or not (False) (default False)
: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).
:param cpnt_type: The type of component to analyse; It \
can be 'nt', 'dnt' or 'aa'.
:param cpnt: The component (nt, aa, dnt) of interest
@@ -340,8 +344,8 @@ def get_stat_cpnt_communities(df: pd.DataFrame, project: str, weight: int,
logging.debug(f"{test_type} for {project}, w:{weight}, "
f"g:{global_weight} cpnt: {cpnt}({cpnt_type})")
outfile, outfile_ctrl = create_outfiles(project, weight, global_weight,
- same_gene, inflation, feature,
- cpnt_type,
+ same_gene, inflation, cell_line,
+ feature, cpnt_type,
cpnt, test_type, community_file)
res = {"project": project, "cpnt": cpnt,
'pval': lm_maker(df, outfile, cpnt)}
@@ -352,7 +356,8 @@ def get_stat_cpnt_communities(df: pd.DataFrame, project: str, weight: int,
def create_dataframe(project: str, weight: int, global_weight: int,
- same_gene: bool, inflation: float, feature: str = 'exon',
+ same_gene: bool, inflation: float, cell_line: str = "ALL",
+ feature: str = 'exon',
region: str = "", component_type: str = 'nt',
community_file: str = "",
from_communities: bool = True,
@@ -367,6 +372,8 @@ def create_dataframe(project: str, weight: int, global_weight: int,
:param same_gene: Say if we consider as co-localised, exons within the \
same gene (True) or not (False)
: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).
:param feature: The kind of feature to analyse
:param from_communities: True if we only select gene/exons
:param region: the region of interest to extract from gene
@@ -385,6 +392,7 @@ def create_dataframe(project: str, weight: int, global_weight: int,
result = ConfigGraph.get_community_file(project, weight,
global_weight,
same_gene, inflation,
+ cell_line,
feature, ".txt")
else:
result = Path(community_file)
@@ -407,7 +415,7 @@ def create_dataframe(project: str, weight: int, global_weight: int,
def create_dataframes(project, weight, global_weight, same_gene, inflation,
- feature,
+ cell_line, feature,
region, test_type, component_type: str,
community_file: str
) -> Tuple[pd.DataFrame, Dict]:
@@ -420,6 +428,8 @@ def create_dataframes(project, weight, global_weight, same_gene, inflation,
: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).
:param feature: The kind of analysed feature
:param region: the region of interest to extract from gene
:param test_type: The type of test to make (permutation or lm)
@@ -430,9 +440,11 @@ def create_dataframes(project, weight, global_weight, same_gene, inflation,
used to find the community files computed with ChIA-PET data.
"""
df = create_dataframe(project, weight, global_weight, same_gene, inflation,
- feature, region, component_type, community_file)
+ cell_line, feature, region, component_type,
+ community_file)
df_ctrl = create_dataframe(project, weight, global_weight, same_gene,
- inflation, feature, region, component_type,
+ inflation, cell_line, feature, region,
+ component_type,
from_communities=False)
df_ctrl = df_ctrl.loc[-df_ctrl[f"id_{feature}"].isin(df[f"id_{feature}"]),
:].copy()
@@ -448,6 +460,7 @@ def multiple_nt_lm_launcher(ps: int,
project: str,
same_gene: bool,
inflation: float,
+ cell_line: str = "ALL",
feature: str = 'exon',
region: str = '',
component_type: str = "nt",
@@ -468,6 +481,8 @@ def multiple_nt_lm_launcher(ps: int,
: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).
:param feature: The kind of analysed feature
:param component_type: The type of component to analyse; It \
can be 'nt', 'dnt' or 'aa'.
@@ -498,16 +513,17 @@ def multiple_nt_lm_launcher(ps: int,
pool = mp.Pool(processes=min(ps, len(condition)))
logging.debug("Creating tables")
df, dic_com = create_dataframes(project, weight, global_weight,
- same_gene, inflation, feature, region,
+ same_gene, inflation, cell_line,
+ feature, region,
test_type, component_type, community_file)
for project, weight, cpnt in condition:
nfile_table = ConfigGraph.get_community_file(
- project, weight, global_weight, same_gene, inflation, feature,
- f"_{component_type}_table.txt", "community_enrichment")
+ project, weight, global_weight, same_gene, inflation, cell_line,
+ feature, f"_{component_type}_table.txt", "community_enrichment")
df.to_csv(nfile_table, sep="\t", index=False)
args = [df, project, weight, global_weight, same_gene, inflation,
- component_type, cpnt, dic_com, feature, test_type, iteration,
- display_size, community_file]
+ cell_line, component_type, cpnt, dic_com, feature, test_type,
+ iteration, display_size, community_file]
processes.append(pool.apply_async(get_stat_cpnt_communities, args))
results = [p.get(timeout=None) for p in processes]
pool.close()
@@ -516,7 +532,8 @@ def multiple_nt_lm_launcher(ps: int,
fdf["padj"] = multipletests(fdf['pval'].values, method='fdr_bh')[1]
outfile = ConfigGraph.get_community_file(project, weight,
global_weight,
- same_gene, inflation, feature,
+ same_gene, inflation, cell_line,
+ feature,
f"lmm-{component_type}_stat.txt",
"community_enrichment")
nfolder = outfile.parent / f"{component_type}_analysis"
@@ -535,6 +552,7 @@ def launcher_community_file(ps: int = 1,
project: str = "GSM1018963_GSM1018964",
same_gene: bool = True,
inflation: float = 1.5,
+ cell_line: str = "ALL",
feature: str = 'exon',
region: str = '',
component_type: str = "nt",
@@ -554,6 +572,8 @@ def launcher_community_file(ps: int = 1,
:param project: The project name, used only if global_weight = 0
:param same_gene: Say if we consider as co-localised exon within the \
same gene (default True)
+ :param cell_line: Interactions are only selected from projects made on \
+ a specific cell line (ALL to disable this filter).
:param feature: The kind of analysed feature (default exon)
:param component_type: The type of component to analyse; It \
can be 'nt', 'dnt' or 'aa'.
@@ -568,7 +588,7 @@ def launcher_community_file(ps: int = 1,
:param logging_level: Level of information to display (default DISABLE)
"""
multiple_nt_lm_launcher(ps, weight, global_weight, project,
- same_gene, inflation, feature, region,
+ same_gene, inflation, cell_line, feature, region,
component_type, test_type, iteration, display_size,
community_file, logging_level)
diff --git a/src/find_interaction_cluster/sf_and_communities.py b/src/find_interaction_cluster/sf_and_communities.py
index 0449bbf88e70be9456a96cb6870f597460e562d4..a9209bcbf43ff7c8ae1fdae21ec92fbc8b36c5c4 100644
--- a/src/find_interaction_cluster/sf_and_communities.py
+++ b/src/find_interaction_cluster/sf_and_communities.py
@@ -181,7 +181,7 @@ def glmm_maker(expanded_df: pd.DataFrame, outfile: Path) -> float:
... "%reg in community": [40, 42.85], 'pval': [1, 0.5], 'padj': [1, 1]})
>>> e_df = expand_dataframe(d)
>>> outfile = ConfigGraph.get_community_file("Test", 1, 1, True, 1.5,
- ... "_stat.txt", "sf_community_enrichment")
+ ... "ALL", "_stat.txt", "sf_community_enrichment")
>>> glmm_maker(e_df, outfile)
1.0
"""
@@ -210,7 +210,7 @@ def glmm_maker(expanded_df: pd.DataFrame, outfile: Path) -> float:
def glmm_statistics(df: pd.DataFrame, sf_name: str, reg: str,
project: str, weight: int, global_weight: int,
- same_gene: bool, inflation: float,
+ same_gene: bool, inflation: float, cell_line: str = "ALL",
feature: str = "exon") -> pd.Series:
"""
Create the glmm statistics for a given splicing factor with \
@@ -229,13 +229,16 @@ def glmm_statistics(df: pd.DataFrame, sf_name: str, reg: str,
same gene (True) or not (False) (default False)
:param feature: The kind of feature analysed
: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).
:return: The glmm pvalue among with other informations
"""
ndf = df.loc[-df['community'].isin(["All-community", "FASTERDB"]),
:].copy()
expanded_df = expand_dataframe(ndf)
outfile = ConfigGraph.get_community_file(project, weight, global_weight,
- same_gene, inflation, feature,
+ same_gene, inflation, cell_line,
+ feature,
f"{sf_name}_{reg}_stat.txt",
"sf_community_enrichment")
noutfold = outfile.parent / "expanded_df"
@@ -276,7 +279,7 @@ def adapt_regulated_list(cnx: sqlite3.Connection,
def get_stat4communities(sf_name: str, reg: str,
project: str, weight: int,
global_weight: int, same_gene: bool,
- inflation: float,
+ inflation: float, cell_line: str = "ALL",
feature: str = 'exon',
) -> Tuple[pd.DataFrame, pd.Series]:
"""
@@ -294,14 +297,16 @@ def get_stat4communities(sf_name: str, reg: str,
:param same_gene: Say if we consider as co-localised, exons within the \
same gene (True) or not (False) (default False)
: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).
:param feature: The kind of analysed feature
"""
logging.debug(f"Working on {sf_name}-{reg}, for {project}, w:{weight}, "
f"g:{global_weight}")
cnx = sqlite3.connect(ConfigGraph.db_file)
result = ConfigGraph.get_community_file(project, weight, global_weight,
- same_gene, inflation, feature,
- ".txt")
+ same_gene, inflation, cell_line,
+ feature, ".txt")
communities = get_communities_basefile(result, 0)
regulated_dic, number = get_every_events_4_a_sl(cnx, sf_name, reg)
reg_ft = regulated_dic[sf_name + "_" + reg]
@@ -332,7 +337,7 @@ def get_stat4communities(sf_name: str, reg: str,
d['project'] = [project] * len(d["community"])
df = pd.DataFrame(d)
s = glmm_statistics(df, sf_name, reg, project, weight, global_weight,
- same_gene, inflation, feature)
+ same_gene, inflation, cell_line, feature)
return df, s
@@ -370,6 +375,7 @@ def multiple_stat_launcher(ps: int,
global_weight: int,
project: str,
same_gene: bool, inflation: float,
+ cell_line: str = "ALL",
feature: str = 'exon',
logging_level: str = "DISABLE"):
"""
@@ -385,6 +391,8 @@ def multiple_stat_launcher(ps: int,
same gene
:param feature: The feature we want to analyse
: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).
:param logging_level: Level of information to display
"""
ConfigGraph.community_folder.mkdir(exist_ok=True, parents=True)
@@ -400,7 +408,7 @@ def multiple_stat_launcher(ps: int,
for project, weight, sf_name, reg in condition:
ckey = get_key(project, weight)
args = [sf_name, reg, project, weight, global_weight, same_gene,
- inflation, feature]
+ inflation, cell_line, feature]
if ckey in processes:
processes[ckey].append(pool.apply_async(get_stat4communities, args))
else:
@@ -415,7 +423,9 @@ def multiple_stat_launcher(ps: int,
df = pd.concat(list_df, axis=0, ignore_index=True)
outfile = ConfigGraph.get_community_file(project, weight,
global_weight,
- same_gene, inflation, feature,
+ same_gene, inflation,
+ cell_line,
+ feature,
"_stat.txt",
"sf_community_enrichment")
df.to_csv(outfile, sep="\t", index=False)
@@ -424,7 +434,9 @@ def multiple_stat_launcher(ps: int,
method='fdr_bh')[1]
outfile = ConfigGraph.get_community_file(project, weight,
global_weight,
- same_gene, inflation, feature,
+ same_gene, inflation,
+ cell_line,
+ feature,
"_glmm_stat.txt",
"sf_community_enrichment")
glm_df.to_csv(outfile, sep="\t", index=False)