diff --git a/src/nt_composition/__main__.py b/src/nt_composition/__main__.py
index 5f4859a4691bd9213c4c43aa3013f8216035e0b3..dd1bb8fa4a62dacbb83b031e61a6935b3b6eeb2f 100644
--- a/src/nt_composition/__main__.py
+++ b/src/nt_composition/__main__.py
@@ -12,21 +12,23 @@ from .make_nt_correlation import create_all_frequency_figures
from .get_projects_interaction import get_interactions_number
import lazyparser as lp
-@lp.parse(weight='weight > 0')
-def launcher(weight: int = 1, global_weight: int = 0,
+@lp.parse(weight='weight > 0', ft_type=['nt', 'dnt', 'tnt', 'codon', 'aa',
+ 'properties'])
+def launcher(weight: int = 1, global_weight: int = 0, ft_type: str = 'nt',
logging_level: str = "DISABLE"):
"""
Launch the creation of density file.
- :param The weight of interaction to consider
+ :param weight: The weight of interaction to consider (default 1)
:param global_weight: The global weight to consider. if \
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 logging_level: The level of data to display
+ by project, else all project are merge together and the interaction \
+ seen in `global_weight` project are taken into account (default 0)
+ :param ft_type: The feature type of interest (default 'nt')
+ :param logging_level: The level of data to display (default 'DISABLE')
"""
get_interactions_number(weight, logging_level)
- create_all_frequency_figures(ConfigNt.cpu, weight, global_weight,
+ create_all_frequency_figures(ConfigNt.cpu, weight, global_weight, ft_type,
logging_level)
diff --git a/src/nt_composition/make_nt_correlation.py b/src/nt_composition/make_nt_correlation.py
index 7234a2625dd46c784d8c79d3599a7cf962c30a94..6283ecb396c08c7f6440fac28261eae4346185d4 100644
--- a/src/nt_composition/make_nt_correlation.py
+++ b/src/nt_composition/make_nt_correlation.py
@@ -102,27 +102,27 @@ def get_all_exon_interacting_with_another(exon: str, arr: np.array
return exons[exons != exon]
-def get_frequency_dic(cnx: sqlite3.Connection, nt: str, ft_type: str
+def get_frequency_dic(cnx: sqlite3.Connection, ft: str, ft_type: str
) -> Dict[str, float]:
"""
- Get the frequency of a nucleotide for every exon in \
+ Get the frequency of a feature for every exon in \
the chia-pet database.
:param cnx: Connection to chia-pet database
- :param nt: The nucleotide of interest
+ :param ft: The feature of interest
:param ft_type: The type of feature of interest
:return: The frequency dic
"""
logging.debug('Calculating frequency table')
query = "SELECT id_exon, frequency " \
"FROM cin_exon_frequency " \
- f"WHERE ft = '{nt}' " \
+ f"WHERE ft = '{ft}' " \
f"AND ft_type = '{ft_type}'"
c = cnx.cursor()
c.execute(query)
result = c.fetchall()
if len(result) == 0:
- msg = f'No Frequencies found for {nt} ({ft_type})'
+ msg = f'No Frequencies found for {ft} ({ft_type})'
logging.exception(msg)
raise NoInteractionError(msg)
dic = {}
@@ -164,10 +164,10 @@ def create_density_table(arr_interaction: np.array, dic_freq: Dict[str, float],
return df
-def create_density_fig(df: pd.DataFrame, project: str, ft_type: str, nt: str,
+def create_density_fig(df: pd.DataFrame, project: str, ft_type: str, ft: str,
weight: int, global_weight: int) -> Tuple[float, float]:
"""
- Compute a density file showing if the nucleotide frequency of \
+ Compute a density file showing if the feature frequency of \
an exons correlates with the frequency in other co-localised exons.
:param df: The dataframe containing frequency of exons and \
@@ -175,7 +175,7 @@ def create_density_fig(df: pd.DataFrame, project: str, ft_type: str, nt: str,
:param project: The name of the project where the co-localisation \
where observed
:param ft_type: The type of feature of interest
- :param nt: The nucleotide of interest
+ :param ft: The feature of interest
:param weight: The minimum weight of interaction to consider
:param global_weight: The global weight to consider. if \
the global weight is equal to 0 then then density figure are calculated \
@@ -195,12 +195,12 @@ def create_density_fig(df: pd.DataFrame, project: str, ft_type: str, nt: str,
plt.plot(df.freq_exon, i + s * df.freq_exon, 'r',
label=f'r={round(r,4)}, p={round(p, 4)}')
plt.legend()
- plt.xlabel(f"Freq of {nt} in an exon")
- plt.ylabel(f"Freq of {nt} in co-localized exons")
- plt.title(f'Freq of {nt} of exons and their co-localized partner in '
+ 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}')
plt.savefig(ConfigNt.get_density_file(weight, global_weight, project,
- ft_type, nt, fig=True))
+ ft_type, ft, fig=True))
plt.close()
return r, p
@@ -326,7 +326,7 @@ def combine_dic(list_dic: List[Dict]) -> Dict:
def create_all_frequency_figures(ps: int, weight: int = 1,
- global_weight: int = 0,
+ global_weight: int = 0, ft_type: str = "nt",
logging_level: str = "DISABLE"):
"""
Make density figure for every selected projects.
@@ -337,6 +337,7 @@ def create_all_frequency_figures(ps: int, weight: int = 1,
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 ft_type: The kind of feature to analyse
:param ps: The number of processes to create
"""
logging_def(ConfigNt.interaction, __file__, logging_level)
@@ -349,12 +350,12 @@ def create_all_frequency_figures(ps: int, weight: int = 1,
# with open(ConfigNt.selected_project, 'r') as f:
# projects = f.read().splitlines()
# projects = projects[:-4]
- nt_list = ['A', 'C', 'G', 'T', 'S', 'W']
- param = product(projects, nt_list, ['nt'])
+ ft_list = ConfigNt.get_features(ft_type)
+ param = product(projects, ft_list, [ft_type])
pool = mp.Pool(processes=ps)
processes = []
- for project, nt, ft_type in param:
- args = [di, project, ft_type, nt, weight, global_weight]
+ for project, ft, ft_type in param:
+ args = [di, project, ft_type, ft, weight, global_weight]
processes.append(pool.apply_async(execute_density_figure_function,
args))
results = []
@@ -366,8 +367,8 @@ def create_all_frequency_figures(ps: int, weight: int = 1,
'ALL', fig=False),
sep="\t")
if global_weight == 0:
- for nt in nt_list:
- create_scatterplot(df_corr, "nt", nt, weight, global_weight)
+ for ft in ft_list:
+ create_scatterplot(df_corr, ft_type, ft, weight, global_weight)
if __name__ == "__main__":