Skip to content

Commit

Permalink
in process
Browse files Browse the repository at this point in the history
  • Loading branch information
@dprada
dprada committed Feb 14, 2024
1 parent 33f3d1e commit e21354a
Show file tree
Hide file tree
Showing 2 changed files with 114 additions and 175 deletions.
104 changes: 89 additions & 15 deletions molsysmt/native/topology.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -20,7 +20,7 @@ def __init__(self, n_atoms=0):
def _fix_null_values(self):

for column in self:
self[column].fillna(pd.NA, inplace=True)
self[column]=self[column].fillna(pd.NA)


class Groups_DataFrame(pd.DataFrame):
Expand All @@ -39,7 +39,7 @@ def __init__(self, n_groups=0):
def _fix_null_values(self):

for column in self:
self[column].fillna(pd.NA, inplace=True)
self[column]=self[column].fillna(pd.NA)


class Components_DataFrame(pd.DataFrame):
Expand All @@ -58,7 +58,7 @@ def __init__(self, n_components=0):
def _fix_null_values(self):

for column in self:
self[column].fillna(pd.NA, inplace=True)
self[column]=self[column].fillna(pd.NA)


class Molecules_DataFrame(pd.DataFrame):
Expand All @@ -77,7 +77,7 @@ def __init__(self, n_molecules=0):
def _fix_null_values(self):

for column in self:
self[column].fillna(pd.NA, inplace=True)
self[column]=self[column].fillna(pd.NA)


class Entities_DataFrame(pd.DataFrame):
Expand All @@ -95,7 +95,7 @@ def __init__(self, n_entities=0):
def _fix_null_values(self):

for column in self:
self[column].fillna(pd.NA, inplace=True)
self[column]=self[column].fillna(pd.NA)


class Chains_DataFrame(pd.DataFrame):
Expand All @@ -113,7 +113,7 @@ def __init__(self, n_chains=0):
def _fix_null_values(self):

for column in self:
self[column].fillna(pd.NA, inplace=True)
self[column]=self[column].fillna(pd.NA)


class Bonds_DataFrame(pd.DataFrame):
Expand All @@ -133,12 +133,12 @@ def __init__(self, n_bonds=0):
def _fix_null_values(self):

for column in self:
self[column].fillna(pd.NA, inplace=True)
self[column]=self[column].fillna(pd.NA)

def _sort_bonds(self):

self_mask = self['atom1_index'] > self['atom2_index']
self.update(self.loc[self_mask].rename({'atom1_index': 'atom2_index', 'atom2_index': 'atom1_index'}, axis=1))
mask = self['atom1_index'] > self['atom2_index']
self.loc[mask, ['atom1_index', 'atom2_index']] = self.loc[mask, ['atom2_index', 'atom1_index']].values
self.sort_values(by=['atom1_index', 'atom2_index'], inplace=True)
self.reset_index(drop=True, inplace=True)

Expand Down Expand Up @@ -344,15 +344,89 @@ def rebuild_components(self, redefine_indices=True, redefine_ids=True, redefine_

self.components["component_id"] = np.arange(self.components.shape[0], dtype=int)

#if redefine_names:
if redefine_types:

from molsysmt.element.component.get_component_type import _get_component_type_from_group_names_and_types

aux_df = self.groups.groupby('component_index').agg(group_name=('group_name', list),
group_type=('group_type', list))
for row in aux_df.itertuples(index=True):
component_type = _get_component_type_from_group_names_and_types(row.group_name, row.group_type)
self.components.iloc[row.Index,2] = component_type

if redefine_names:

from molsysmt.element.group.small_molecule import small_molecule_names

aux_df = self.groups.groupby('component_index').agg(group_name=('group_name', list),
group_type=('group_type', list))

component_types = self.components['component_type'].to_numpy()

counter = {'peptide':0, 'protein':0, 'small molecule':0, 'unknown':0}

peptides = {}
proteins = {}
small_molecules = {}

for component_type, row in zip(component_types, aux_df.itertuples(index=True)):

if component_type == 'peptide':

string_peptide = ','.join(row.group_name)

if string_peptide in peptides:
component_name = peptides[string_peptide]
else:
component_name = component_type+' '+str(counter[component_type])
peptides[string_peptide] = component_name
counter[component_type] += 1

elif component_type == 'protein':

string_protein = ','.join(row.group_name)

if string_protein in proteins:
component_name = proteins[string_protein]
else:
component_name = component_type+' '+str(counter[component_type])
proteins[string_protein] = component_name
counter[component_type] += 1

elif component_type == 'small molecule':

group_name = row.group_name[0]

if group_name in small_molecules:
component_name = small_molecules[group_name]
else:
if group_name in small_molecule_names:
component_name = small_molecule_names[group_name]
else:
component_name = group_name
small_molecules[component_name] = component_name

elif component_type in ['ion', 'lipid']:

component_name = row.group_name[0]

elif component_type in ['water']:

component_name = 'water'

else:

component_name = 'unknown '+str(counter['unknown'])
counter['unknown']+=1

self.components.iloc[row.Index,1] = component_name



# component_name = get_component_name(self, element='component', redefine_names=True, skip_digestion=True)
# self.components["component_name"] = np.array(component_name, dtype=object)
# del component_name

#if redefine_types:
# component_type = get_component_type(self, element='component', redefine_types=True, skip_digestion=True)
# self.components["component_type"] = np.array(component_type, dtype=object)
# del component_type


def rebuild_molecules(self, redefine_indices=True, redefine_ids=True, redefine_names=True, redefine_types=True):
Expand Down Expand Up @@ -464,7 +538,7 @@ def rebuild_entities(self, redefine_indices=True, redefine_ids=True, redefine_na
else:
entity_index = aux_dict[molecule_name]
else:
if 'unknown' in aux_dict:
if 'unknown' not in aux_dict:
aux_dict['unknown'] = count
entity_index = count
count += 1
Expand Down
Loading

0 comments on commit e21354a

Please sign in to comment.