dcatoolkit 0.1.8__tar.gz → 0.2.0__tar.gz

{dcatoolkit-0.1.8/src/dcatoolkit.egg-info → dcatoolkit-0.2.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: dcatoolkit
-Version: 0.1.8
+Version: 0.2.0
 Summary: Collection of useful modules and representations for managing DCA output data.
 Author-email: Raheel Syed Ahmed <raheelsyedahmed@gmail.com>
 Maintainer-email: Raheel Syed Ahmed <raheelsyedahmed@gmail.com>
@@ -38,7 +38,7 @@ Classifier: Programming Language :: Python :: 3.12
 Requires-Python: >=3.10
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: biotite
+Requires-Dist: biotite>=1.0.1
 Requires-Dist: matplotlib>=3.8.0
 Requires-Dist: numpy>=1.26.0
 Requires-Dist: pandas>=2.1.0

{dcatoolkit-0.1.8 → dcatoolkit-0.2.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "dcatoolkit"
-version = "0.1.8"
+version = "0.2.0"
 description = "Collection of useful modules and representations for managing DCA output data."
 keywords = ["dca", "toolkit", "DI", "coevolution"]
 
@@ -21,7 +21,7 @@ maintainers = [
 ]
 
 dependencies = [
-  "biotite",
+  "biotite>=1.0.1",
   "matplotlib>=3.8.0",
   "numpy>=1.26.0",
   "pandas>=2.1.0",

dcatoolkit-0.2.0/src/dcatoolkit/__init__.py

@@ -0,0 +1,6 @@
+
+__version__ = "0.1.9"
+from .representation import Pairs, DirectInformationData, StructureInformation, ResidueAlignment, MMCIFInformation, PDBInformation
+from .analytics import MSATools
+
+__all__ = ['Pairs', 'DirectInformationData', 'StructureInformation', 'ResidueAlignment', 'MMCIFInformation', 'PDBInformation', 'MSATools']

{dcatoolkit-0.1.8 → dcatoolkit-0.2.0}/src/dcatoolkit/analytics.py

@@ -1,7 +1,8 @@
 import re
 from collections import Counter
 from typing import Optional, Union
-import string, io
+import string
+import io
 
 class MSATools:
     """
@@ -107,7 +108,7 @@ class MSATools:
             List of entries that are valid in that their sequences' number of maximum continuous gaps is within the threshold supplied as `max_gaps`. 
         """
         table = str.maketrans('', '', string.ascii_lowercase+".")
-        if max_gaps == None:
+        if max_gaps is None:
             kept_entries = []
             for header, sequence in self.MSA:
                 sequence = sequence.translate(table)

{dcatoolkit-0.1.8 → dcatoolkit-0.2.0}/src/dcatoolkit/representation.py

@@ -4,10 +4,11 @@ from scipy.spatial.distance import cdist
 
 import biotite.structure as struc
 import biotite.structure.io.pdbx as pdbx
+import biotite.structure.io.pdb as pdb
 import biotite.database.rcsb as rcsb
 
 from collections.abc import Iterable
-from typing import Optional, Union
+from typing import Optional, Union, Literal, overload
 import numpy.typing as npt
 
 
@@ -619,9 +620,124 @@ class StructureInformation:
     """
     Information regarding a protein structure, obtained from a protein structure file.
 
+    Uses fetch_pdb() to pull protein structure information from RCSB. Uses read_x_file() to supply a filepath to pull protein structure information from a file.
+    """
+    @overload
+    @staticmethod
+    def fetch_pdb(pdb_id: str, struc_format: Literal["mmcif"]="mmcif", model_num: int=1) -> 'MMCIFInformation':
+        ...
+
+    @overload
+    @staticmethod
+    def fetch_pdb(pdb_id: str, struc_format: Literal["pdb"], model_num: int=1) -> 'PDBInformation':
+        ...
+
+
+    @staticmethod
+    def fetch_pdb(pdb_id: str, struc_format: Literal["mmcif", "pdb"]="mmcif", model_num: int=1) -> Union['MMCIFInformation', 'PDBInformation']:
+        """
+        Fetches PDB as mmCIF file from RCSB and compiles the information into a StructureInformation instance.
+
+        Parameters
+        ----------
+        pdb_id : str
+            PDB ID to be fetched from the RCSB database.
+        struc_format : str
+            The format of the file to pull from the RCSB database.
+        model_num : int
+            The model number to access from the PDB to ensure an AtomArray is returned containing the atom information of the protein structure.
+            
+        Returns
+        -------
+        StructureInformation
+            StructureInformation generated from pdbx.get_structure() function using the pdbx file fetched from RCSB.
+        
+        Raises
+        ------
+        TypeError
+            Fetched data was not found and returned None instead.
+        ValueError
+            Structure format may be invalid (not PDBx/mmCIF or PDB).
+        """
+        fetched_data = rcsb.fetch(pdb_id, struc_format)
+        if fetched_data is None:
+            raise TypeError("RCSB fetch failed. Try fetch again.")
+        elif struc_format == "mmcif":
+            pdbx_file = pdbx.CIFFile.read(fetched_data)
+            return MMCIFInformation(pdbx.get_structure(pdbx_file=pdbx_file, model=model_num, use_author_fields=False), pdbx_file, model_num)
+        elif struc_format == "pdb":
+            pdb_file = pdb.PDBFile.read(fetched_data)
+            return PDBInformation(pdb.get_structure(pdb_file=pdb_file, model=model_num), pdb_file=pdb_file, model_num=model_num)
+        else:
+            raise ValueError(f"struc_format {struc_format} is not valid or currently supported by DCA Toolkit")
+    @staticmethod
+    def read_mmCIF_file(pdbx_filepath: str, model_num: int=1) -> 'MMCIFInformation':
+        """
+        Reads PDB mmCIF file from filepath and compiles the information into a CIFInformation instance.
+
+        Parameters
+        ----------
+        pdbx_filepath : str
+            Filepath of the PDB mmCIF file to be read.
+        model_num : int
+            The model number to access from the PDB to ensure an AtomArray is returned containing the atom information of the protein structure.
+
+        Returns
+        -------
+        CIFInformation
+            CIFInformation generated from pdbx.get_structure() function using the PDBx file read from the pdbx_filepath.
+        """
+        pdbx_file = pdbx.CIFFile.read(pdbx_filepath)
+        return MMCIFInformation(pdbx.get_structure(pdbx_file, model=model_num, use_author_fields=False), pdbx_file, model_num)
+    
+    @staticmethod
+    def read_pdb_file(pdb_filepath: str, model_num: int=1) -> 'PDBInformation':
+        """
+        Reads PDB file from filepath and compiles the information into a PDBInformation instance.
+
+        Parameters
+        ----------
+        pdb_filepath : str
+            Filepath of the PDB mmCIF file to be read.
+        model_num : int
+            The model number to access from the PDB to ensure an AtomArray is returned containing the atom information of the protein structure.
+
+        Returns
+        -------
+        PDBInformation
+            PDBInformation generated from pdb.get_structure() function using the PDB file read from the pdb_filepath.
+        """
+        pdb_file = pdb.PDBFile.read(pdb_filepath)
+        return PDBInformation(pdb.get_structure(pdb_file, model=model_num), pdb_file, model_num)
+    
+    @staticmethod
+    def write_contacts_set(filepath : str, contacts_set : set[tuple[int, int]]) -> None:
+        """
+        Write the contacts generated from get_contacts or general set of tuples of pairs.
+
+        Parameters
+        ----------
+        filepath : str
+            Path of file to output contacts_set to.
+        contacts_set : set of tuple of int, int
+            Set of tuples of pairs that represent contacts.
+        
+        Returns
+        -------
+        None
+        """
+        contacts_list = list(sorted(contacts_set))
+        with open(filepath, 'w') as fs:
+            for pair in contacts_list:
+                fs.write(str(pair[0]) + "\t" + str(pair[1]) + "\n")
+
+class MMCIFInformation(StructureInformation):
+    """
+    Information regarding a protein structure, obtained from a protein structure file.
+
     Parameters
     ----------
-    structure : biotite.structure
+    structure : biotite.structure.AtomArray
         Structure obtained from an RCSB entry with a provided pdbx/mmcif file with a specified model number.
     pdbx_file : biotite.io.pdbx.CIFFile
         mmCIF file that contains generic information and atomic information of the protein structure categorized into mmCIF blocks.
@@ -707,56 +823,6 @@ class StructureInformation:
                     self.res_auth_dict[unique_entry[2]] = unique_entry[[1,3]].astype('int')
         else:
             self.atom_site_category = None
-
-    @staticmethod
-    def fetch_pdb(pdb_id: str, model_num: int=1, struc_format: str="mmcif") -> 'StructureInformation':
-        """
-        Fetches PDB as mmCIF file from RCSB and compiles the information into a StructureInformation instance.
-
-        Parameters
-        ----------
-        pdb_id : str
-            PDB ID to be fetched from the RCSB database.
-        model_num : int
-            The model number to access from the PDB to ensure an AtomArray is returned containing the atom information of the protein structure.
-        struc_format : str
-            The format of the file to pull from the RCSB database.
-        
-        Returns
-        -------
-        StructureInformation
-            StructureInformation generated from pdbx.get_structure() function using the pdbx file fetched from RCSB. The pdbx file is also supplied as an argument.
-        
-        Raises
-        ------
-        TypeError
-            Fetched data was not found and returned None instead.
-        """
-        fetched_data = rcsb.fetch(pdb_id, struc_format)
-        if fetched_data is None:
-            raise TypeError("RCSB fetch failed. Try fetch again.")
-        pdbx_file = pdbx.CIFFile.read(fetched_data)
-        return StructureInformation(pdbx.get_structure(pdbx_file=pdbx_file, model=model_num, use_author_fields=False), pdbx_file, model_num)
-    
-    @staticmethod
-    def read_pdb_mmCIF(pdb_filepath: str, model_num: int=1) -> 'StructureInformation':
-        """
-        Reads PDB mmCIF file from filepath and compiles the information into a StructureInformation instance.
-
-        Parameters
-        ----------
-        pdb_filepath : str
-            Filepath of the PDB mmCIF file to be read.
-        model_num : int
-            The model number to access from the PDB to ensure an AtomArray is returned containing the atom information of the protein structure.
-
-        Returns
-        -------
-        StructureInformation
-            StructureInformation generated from pdbx.get_structure() function using the pdbx file fetched from RCSB. The pdbx file is also supplied as an argument.
-        """
-        pdbx_file = pdbx.CIFFile.read(pdb_filepath)
-        return StructureInformation(pdbx.get_structure(pdbx_file, model=model_num, use_author_fields=False), pdbx_file, model_num)
     
     def get_full_sequence(self, chain_id: str, auth_chain_id_supplied: bool=False) -> str:
         """
@@ -801,7 +867,7 @@ class StructureInformation:
         else:
             return self.non_missing_sequences[chain_id]
         
-    def get_chain_specific_structure(self, ca_only: bool, chain1: str, chain2: str, remove_hetero=True) -> tuple:
+    def get_chain_specific_structure(self, ca_only: bool, chain1: str, chain2: str, remove_hetero=True, auth_chain_id_supplied: bool=False) -> tuple:
         """
         Subsets structure attribute to select for chain specific portions of the structure.
 
@@ -815,12 +881,18 @@ class StructureInformation:
             Chain id corresponding to the second column of residues in the structure.
         remove_hetero : bool, default=True
             If true, the structure will also be subsetted for atom entries where the hetero annotation is False, thus removing heteroatoms.
+        auth_chain_id_supplied : bool
+            If True, the chain_id supplied is the auth chain id found on the RCSB website.
 
         Returns
         -------
         tuple of biotite.structure.AtomArray, biotite.structure.AtomArray
             Two AtomArrays that refer to atoms in the first chain and second chain, respectively without accounting for the presence of heteroatoms if `remove_hetero` is True.
         """
+        if auth_chain_id_supplied:
+            chain1 = self.auth_chain_dict[chain1]
+            chain2 = self.auth_chain_dict[chain2]
+
         selected_structure = self.structure
         if remove_hetero:
             # Remove hetero atoms via hetero column of structure ndarray
@@ -832,7 +904,7 @@ class StructureInformation:
         chain2_structure = selected_structure[selected_structure.chain_id == chain2]
         return (chain1_structure, chain2_structure)
 
-    def generate_dist_matrix(self, ca_only: bool, chain1: str, chain2: str):
+    def generate_dist_matrix(self, ca_only: bool, chain1: str, chain2: str, auth_chain_id_supplied: bool=False):
         """
         Generates distance matrix between two chains in the structure attribute.
 
@@ -844,32 +916,40 @@ class StructureInformation:
             Chain id corresponding to the first column of residues in the structure.
         chain2 : str
             Chain id corresponding to the first column of residues in the structure.
+        auth_chain_id_supplied : bool
+            If True, the chain_id supplied is the auth chain id found on the RCSB website.
 
         Returns
         -------
         tuple of biotite.structure.AtomArray, biotite.structure.AtomArray, numpy.ndarray
             Tuple containing the chain 1 structure, the chain 2 structure, and the distance matrix of chain 1 and chain 2's pairwise distances.
         """
-        chain1_structure, chain2_structure = self.get_chain_specific_structure(ca_only, chain1, chain2, remove_hetero=True)
+        chain1_structure, chain2_structure = self.get_chain_specific_structure(ca_only, chain1, chain2, remove_hetero=True, auth_chain_id_supplied=auth_chain_id_supplied)
         dist_matrix = cdist(chain1_structure.coord, chain2_structure.coord)
         return (chain1_structure, chain2_structure, dist_matrix)
     
-    def get_shift_values(self, chain1: str, chain2: str) -> tuple[int, int]:
+    def get_shift_values(self, chain1: str, chain2: str, auth_chain_id_supplied: bool=False) -> tuple[int, int]:
         """
         Get shift values needed for production of auth residue ids.
         
         Parameters
         ----------
         chain1 : str
-            Name of the chain id present in the struct_ref_seq block of cif files referring to the second column of residues.
+            Name of the chain id present referring to the second column of residues.
         chain2 : str
-            Name of the chain id present in the struct_ref_seq block of cif files referring to the second column of residues.
+            Name of the chain id present referring to the second column of residues.
+        auth_chain_id_supplied : bool
+            If True, the chain_id supplied is the auth chain id found on the RCSB website.
 
         Returns
         -------
         (shift1, shift2) : tuple of int, int
             Tuple containing both shift values, the difference between the auth_res_id and res_id.
         """
+        if auth_chain_id_supplied:
+            chain1 = self.auth_chain_dict[chain1]
+            chain2 = self.auth_chain_dict[chain2]
+
         shift1 = 0
         shift2 = 0
         if self.atom_site_category:
@@ -879,7 +959,7 @@ class StructureInformation:
         else:
             return shift1, shift2
 
-    def get_min_dist_atom_info(self, pairs: npt.NDArray, chain1: str, chain2: str) -> npt.NDArray:
+    def get_min_dist_atom_info(self, pairs: npt.NDArray, chain1: str, chain2: str, auth_chain_id_supplied: bool=False) -> npt.NDArray:
         """
         Generate a ndarray of residue ids and their corresponding atom names such that the distance is the minimum between the initial residues provided.
 
@@ -891,14 +971,16 @@ class StructureInformation:
             Chain id corresponding to the first column of residues in the structure.
         chain2 : str
             Chain id corresponding to the second column of residues in the structure.
-
+        auth_chain_id_supplied : bool
+            If True, the chain_id supplied is the auth chain id found on the RCSB website.
+            
         Returns
         -------
         min_dist_pairs_atoms_arr : numpy.ndarray
-            Structured ndarray that has residue indices, auth residue indices (corresponding to the protein numbering), and atomic names in the format {'names': ['residue1','residue2','auth_residue1','auth_residue2','atom_name1','atom_name2'], 'formats': [int,int,str,str]}
+            Structured ndarray that has residue indices, auth residue indices (corresponding to the protein numbering), and atomic names in the format {'names': ['residue1','residue2','auth_residue1','auth_residue2','atom_name1','atom_name2'], 'formats': [int,int,int,int,'<U10','<U10']}
         """
-        shift1, shift2 = self.get_shift_values(chain1, chain2)
-        chain1_structure, chain2_structure = self.get_chain_specific_structure(ca_only=False, chain1=chain1, chain2=chain2, remove_hetero=True)
+        shift1, shift2 = self.get_shift_values(chain1, chain2, auth_chain_id_supplied=auth_chain_id_supplied)
+        chain1_structure, chain2_structure = self.get_chain_specific_structure(ca_only=False, chain1=chain1, chain2=chain2, remove_hetero=True, auth_chain_id_supplied=auth_chain_id_supplied)
         min_dist_pairs_atoms = []
         for row in pairs:
             # Obtain structure information for chains 1 and 2
@@ -917,7 +999,7 @@ class StructureInformation:
         min_dist_pairs_atoms_arr = np.array(min_dist_pairs_atoms, dtype={'names': ['residue1','residue2','auth_residue1','auth_residue2','atom_name1','atom_name2'], 'formats': [int,int,int,int,'<U10','<U10']})
         return min_dist_pairs_atoms_arr
 
-    def get_contacts(self, ca_only: bool, threshold: float, chain1: str, chain2: str, auth_contacts: bool=False) -> set[tuple[int, int]]:
+    def get_contacts(self, ca_only: bool, threshold: float, chain1: str, chain2: str, auth_contacts: bool=False, auth_chain_id_supplied: bool=False) -> set[tuple[int, int]]:
         """
         Get contacts from the structure attribute where the distance between two residues is less than the threshold.
 
@@ -932,15 +1014,17 @@ class StructureInformation:
         chain2 : str
             Chain id corresponding to the second column of residues in the structure.
         auth_contacts : bool
-            True if supplying alt_ids for residues indices, False if cif residue indexing is needed.
+            True if you want alt_ids for residues indices, False if cif residue indexing is needed.
+        auth_chain_id_supplied : bool
+            If True, the chain_id supplied is the auth chain id found on the RCSB website.
 
         Returns
         -------
         contacts_set : set of tuple of ints
             Set of contacts, tuples with "residue1" and "residue2" from the structure that are within the distance threshold.
         """
-        shift1, shift2 = self.get_shift_values(chain1, chain2)
-        chain1_structure, chain2_structure, dist_matrix = self.generate_dist_matrix(ca_only, chain1, chain2)
+        
+        chain1_structure, chain2_structure, dist_matrix = self.generate_dist_matrix(ca_only, chain1, chain2, auth_chain_id_supplied=auth_chain_id_supplied)
         thresh_ind = np.argwhere(dist_matrix <= threshold)
         contacts_set = set()
         for indices in thresh_ind:
@@ -950,28 +1034,208 @@ class StructureInformation:
             res2 = chain2_atom.res_id
             if not(chain1==chain2 and res1 >= res2):
                 if auth_contacts:
+                    shift1, shift2 = self.get_shift_values(chain1, chain2, auth_chain_id_supplied=auth_chain_id_supplied)
                     contacts_set.add((res1 + shift1, res2 + shift2))
                 else:
                     contacts_set.add((res1, res2))
         return contacts_set
+
+class PDBInformation(StructureInformation):
+    """
+    Information regarding a protein structure, obtained from a protein structure file.
+
+    Parameters
+    ----------
+    structure : biotite.structure.AtomArray
+        Structure obtained from an RCSB entry with a provided pdbx/mmcif file with a specified model number.
+    pdb_file : biotite.io.pdb.PDBFile
+        mmCIF file that contains generic information and atomic information of the protein structure categorized into mmCIF blocks.
+    model_num : int
+        The model number to access from the PDB to ensure an AtomArray is returned containing the atom information of the protein structure.
+
+    Attributes
+    ----------
+    self.non_missing_sequences : dict of str, biotite.sequence.ProteinSequence
+        The protein sequences, without missing residues, compiled in the structure of the StructureInformation instance stored in a dictionary where chain_id is the key and the sequence string is the value.
     
-    @staticmethod
-    def write_contacts_set(filepath : str, contacts_set : set[tuple[int, int]]) -> None:
+    """
+    def __init__(self, structure, pdb_file: pdb.PDBFile, model_num: int):
+        self.structure = structure
+        self.pdb_file = pdb_file
+        self.model_num = model_num
+        non_hetero_structure = self.structure[self.structure.hetero == False]
+        self.non_missing_sequences = {str(chain): str(sequence) for (chain, sequence) in list(zip(struc.get_chains(non_hetero_structure), struc.to_sequence(non_hetero_structure)[0]))}
+        self.unique_chains = struc.get_chains(non_hetero_structure)
+
+    def get_non_missing_sequence(self, chain_id: str) -> str:
         """
-        Write the contacts generated from get_contacts or general set of tuples of pairs.
+        Get sequence, including only non-missing residues, from the specified chain.
 
         Parameters
         ----------
-        filepath : str
-            Path of file to output contacts_set to.
-        contacts_set : set of tuple of int, int
-            Set of tuples of pairs that represent contacts.
+        chain_id : str
+            Chain id supplied. The full sequence, including only non-missing residues, of this chain will be returned.
+
+        Returns
+        -------
+        str
+            The full sequence, including only non-missing residues, of the chain specified.
+        """
+        return self.non_missing_sequences[chain_id]
+    
+    def get_chain_specific_structure(self, ca_only: bool, chain1: str, chain2: str, remove_hetero=True) -> tuple:
+        """
+        Subsets structure attribute to select for chain specific portions of the structure.
+
+        Parameters
+        ----------
+        ca_only : bool
+            If true, the structure will also be subsetted for atom entries where the atom_name annotation is "CA" (referring to alpha-carbons)
+        chain1 : str
+            Chain id corresponding to the first column of residues in the structure.
+        chain2 : str
+            Chain id corresponding to the second column of residues in the structure.
+        remove_hetero : bool, default=True
+            If true, the structure will also be subsetted for atom entries where the hetero annotation is False, thus removing heteroatoms.
+
+        Returns
+        -------
+        tuple of biotite.structure.AtomArray, biotite.structure.AtomArray
+            Two AtomArrays that refer to atoms in the first chain and second chain, respectively without accounting for the presence of heteroatoms if `remove_hetero` is True.
+        """
+
+        selected_structure = self.structure
+        if remove_hetero:
+            # Remove hetero atoms via hetero column of structure ndarray
+            selected_structure = self.structure[self.structure.hetero == False]
+        if ca_only:
+            # Consider selection of alpha-carbon atoms only
+            selected_structure = selected_structure[selected_structure.atom_name == "CA"]
+        chain1_structure = selected_structure[selected_structure.chain_id == chain1]
+        chain2_structure = selected_structure[selected_structure.chain_id == chain2]
+        return (chain1_structure, chain2_structure)
+    
+    def generate_dist_matrix(self, ca_only: bool, chain1: str, chain2: str):
+        """
+        Generates distance matrix between two chains in the structure attribute.
+
+        Parameters
+        ----------
+        ca_only : bool
+            If True, only atoms that have the name "CA" are selected in the chains the distance matrix is calculated between.
+        chain1 : str
+            Chain id corresponding to the first column of residues in the structure.
+        chain2 : str
+            Chain id corresponding to the first column of residues in the structure.
+
+        Returns
+        -------
+        tuple of biotite.structure.AtomArray, biotite.structure.AtomArray, numpy.ndarray
+            Tuple containing the chain 1 structure, the chain 2 structure, and the distance matrix of chain 1 and chain 2's pairwise distances.
+        """
+        chain1_structure, chain2_structure = self.get_chain_specific_structure(ca_only, chain1, chain2, remove_hetero=True)
+        dist_matrix = cdist(chain1_structure.coord, chain2_structure.coord)
+        return (chain1_structure, chain2_structure, dist_matrix)
+
+    def get_shift_values(self, chain1: str, chain2: str) -> tuple[int, int]:
+        """
+        Get shift values needed for production of auth residue ids.
         
+        Parameters
+        ----------
+        chain1 : str
+            Name of the chain id present in the struct_ref_seq block of cif files referring to the second column of residues.
+        chain2 : str
+            Name of the chain id present in the struct_ref_seq block of cif files referring to the second column of residues.
+
         Returns
         -------
-        None
+        (shift1, shift2) : tuple of int, int
+            Tuple containing both shift values, the difference between the auth_res_id and res_id.
         """
-        contacts_list = list(sorted(contacts_set))
-        with open(filepath, 'w') as fs:
-            for pair in contacts_list:
-                fs.write(str(pair[0]) + "\t" + str(pair[1]) + "\n")
+        non_hetero_structure = self.structure[self.structure.hetero == False]
+        shift1 = 0
+        shift2 = 0
+        if chain1 in self.unique_chains and chain2 in self.unique_chains:
+            shift1 = non_hetero_structure[non_hetero_structure.chain_id == chain1][0].res_id - 1
+            shift2 = non_hetero_structure[non_hetero_structure.chain_id == chain2][0].res_id - 1
+            return shift1, shift2
+        else:
+            return shift1, shift2
+
+    def get_min_dist_atom_info(self, pairs: npt.NDArray, chain1: str, chain2: str) -> npt.NDArray:
+        """
+        Generate a ndarray of residue ids and their corresponding atom names such that the distance is the minimum between the initial residues provided.
+
+        Parameters
+        ----------
+        pairs : numpy.ndarray
+            Pairs structured ndarray with "residue1" and "residue2" columns.
+        chain1 : str
+            Chain id corresponding to the first column of residues in the structure.
+        chain2 : str
+            Chain id corresponding to the second column of residues in the structure.
+            
+        Returns
+        -------
+        min_dist_pairs_atoms_arr : numpy.ndarray
+            Structured ndarray that has residue indices, auth residue indices (corresponding to the protein numbering), and atomic names in the format {'names': ['residue1','residue2','auth_residue1','auth_residue2','atom_name1','atom_name2'], 'formats': [int,int,int,int,'<U10','<U10']}
+        """
+        shift1, shift2 = self.get_shift_values(chain1, chain2)
+        chain1_structure, chain2_structure = self.get_chain_specific_structure(ca_only=False, chain1=chain1, chain2=chain2, remove_hetero=True)
+        min_dist_pairs_atoms = []
+        for row in pairs:
+            # Obtain structure information for chains 1 and 2
+            chain1_res1_structure = chain1_structure[chain1_structure.res_id == row['residue1']]
+            chain2_res2_structure = chain2_structure[chain2_structure.res_id == row['residue2']]
+            
+            # Calculate a distance matrix and find the indices of the minimal value in the matrix
+            dist_matrix = cdist(chain1_res1_structure.coord, chain2_res2_structure.coord)
+            
+            ind = np.unravel_index(np.argmin(dist_matrix), dist_matrix.shape)
+            # Use the indices to access the atom in the atom array and get the correct atom name.
+            # Generate the auth ids of the residues in the pairs ndarray
+            orig_res_id1 = row['residue1'] - shift1
+            orig_res_id2 = row['residue2'] - shift2
+            min_dist_pairs_atoms.append((orig_res_id1, orig_res_id2, row['residue1'], row['residue2'], chain1_res1_structure[ind[0]].atom_name, chain2_res2_structure[ind[1]].atom_name))
+        min_dist_pairs_atoms_arr = np.array(min_dist_pairs_atoms, dtype={'names': ['residue1','residue2','auth_residue1','auth_residue2','atom_name1','atom_name2'], 'formats': [int,int,int,int,'<U10','<U10']})
+        return min_dist_pairs_atoms_arr    
+
+    def get_contacts(self, ca_only: bool, threshold: float, chain1: str, chain2: str, auth_contacts: bool=False) -> set[tuple[int, int]]:
+        """
+        Get contacts from the structure attribute where the distance between two residues is less than the threshold.
+
+        Parameters
+        ----------
+        ca_only : bool
+            If true, only consider alpha-carbon to alpha-carbon distances. 
+        threshold : float
+            Maximum distance to consider between two atoms.
+        chain1 : str
+            Chain id corresponding to the first column of residues in the structure.
+        chain2 : str
+            Chain id corresponding to the second column of residues in the structure.
+        auth_contacts : bool
+            True if you want alt_ids for residues indices, False if cif residue indexing is needed.
+
+        Returns
+        -------
+        contacts_set : set of tuple of ints
+            Set of contacts, tuples with "residue1" and "residue2" from the structure that are within the distance threshold.
+        """
+        
+        chain1_structure, chain2_structure, dist_matrix = self.generate_dist_matrix(ca_only, chain1, chain2)
+        thresh_ind = np.argwhere(dist_matrix <= threshold)
+        contacts_set = set()
+        for indices in thresh_ind:
+            chain1_atom = chain1_structure[indices[0]]
+            chain2_atom = chain2_structure[indices[1]]
+            res1 = chain1_atom.res_id
+            res2 = chain2_atom.res_id
+            if not(chain1==chain2 and res1 >= res2):
+                if not auth_contacts:
+                    shift1, shift2 = self.get_shift_values(chain1, chain2)
+                    contacts_set.add((res1 - shift1, res2 - shift2))
+                else:
+                    contacts_set.add((res1, res2))
+        return contacts_set

{dcatoolkit-0.1.8 → dcatoolkit-0.2.0/src/dcatoolkit.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: dcatoolkit
-Version: 0.1.8
+Version: 0.2.0
 Summary: Collection of useful modules and representations for managing DCA output data.
 Author-email: Raheel Syed Ahmed <raheelsyedahmed@gmail.com>
 Maintainer-email: Raheel Syed Ahmed <raheelsyedahmed@gmail.com>
@@ -38,7 +38,7 @@ Classifier: Programming Language :: Python :: 3.12
 Requires-Python: >=3.10
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: biotite
+Requires-Dist: biotite>=1.0.1
 Requires-Dist: matplotlib>=3.8.0
 Requires-Dist: numpy>=1.26.0
 Requires-Dist: pandas>=2.1.0

{dcatoolkit-0.1.8 → dcatoolkit-0.2.0}/src/dcatoolkit.egg-info/SOURCES.txt

@@ -8,4 +8,5 @@ src/dcatoolkit.egg-info/PKG-INFO
 src/dcatoolkit.egg-info/SOURCES.txt
 src/dcatoolkit.egg-info/dependency_links.txt
 src/dcatoolkit.egg-info/requires.txt
-src/dcatoolkit.egg-info/top_level.txt
+src/dcatoolkit.egg-info/top_level.txt
+tests/test_contacts.py

{dcatoolkit-0.1.8 → dcatoolkit-0.2.0}/src/dcatoolkit.egg-info/requires.txt

@@ -1,4 +1,4 @@
-biotite
+biotite>=1.0.1
 matplotlib>=3.8.0
 numpy>=1.26.0
 pandas>=2.1.0

dcatoolkit-0.2.0/tests/test_contacts.py

@@ -0,0 +1,95 @@
+from context import MMCIFInformation, PDBInformation
+from pathlib import Path
+import biotite.structure.io.pdbx as pdbx
+import biotite.database.rcsb as rcsb
+
+pdb_id_chain_map = {}
+with open("tests/pdb_info/pdb_ids.txt", "r") as fs:
+    pdb_id_data = fs.read().splitlines()
+    for line in pdb_id_data:
+        pdb_id, chain1, chain2 = line.split()
+        if len(chain1.split("/")) > 1:
+            chain1, auth_chain1 = chain1.split("/")
+        else:
+            auth_chain1 = chain1
+        if len(chain2.split("/")) > 1:
+            chain2, auth_chain2 = chain2.split("/")
+        else:
+            auth_chain2 = chain2
+        pdb_id_chain_map[pdb_id] = (chain1, auth_chain1, chain2, auth_chain2)        
+
+def read_contacts(input_filepath: str) -> set[tuple[int, int]]:
+    with open(input_filepath, 'r') as fs:
+        data = fs.read().splitlines()
+    results = set()
+    for pair in data:
+        res1, res2 = pair.split()
+        res1 = int(res1)
+        res2 = int(res2)
+        if res1 < res2:
+            results.add((int(res1), int(res2)))
+    return results
+    #return set([(int(x.split()[0]), int(x.split()[1])) for x in data])
+
+def drop_inord_res(contacts: set[tuple[int, int]]) -> set[tuple[int, int]]:
+    results = set()
+    for contact in contacts:
+        if contact[0] < contact[1]:
+            results.add(contact)
+    return results
+
+def compare(*args) -> None:
+    for arg1 in args:
+        for arg2 in args:
+            assert arg1 == arg2
+
+def check_contacts(test_CA: bool, threshold: float):
+    cif_files = list(Path("tests/pdb_info").glob("*.cif"))
+    if test_CA:
+        print("Testing CA")
+    else:
+        print("Testing AA")
+
+    for cif_file in cif_files:
+        pdb_id = cif_file.stem.upper()
+        if test_CA:
+            corresponding_file = f"tests/pdb_info/monomer_calpha_{pdb_id}_10.0"
+        else:
+            corresponding_file = f"tests/pdb_info/monomer_allatom_{pdb_id}_8.0"
+        
+        cif_file_contacts = read_contacts(corresponding_file)
+        chain1, auth_chain1, chain2, auth_chain2 = pdb_id_chain_map[pdb_id]
+        fetch_cif_contacts = {(int(x[0]), int(x[1])) for x in MMCIFInformation.fetch_pdb(pdb_id).get_contacts(test_CA, threshold, chain1, chain2, auth_contacts=True)}
+        read_cif_contacts = {(int(x[0]), int(x[1])) for x in MMCIFInformation.read_mmCIF_file(str(cif_file)).get_contacts(test_CA, threshold, chain1, chain2, auth_contacts=True)}
+        fetch_authchain_cif_contacts = {(int(x[0]), int(x[1])) for x in MMCIFInformation.fetch_pdb(pdb_id).get_contacts(test_CA, threshold, auth_chain1, auth_chain2, auth_contacts=True, auth_chain_id_supplied=True)}
+        read_authchain_cif_contacts = {(int(x[0]), int(x[1])) for x in MMCIFInformation.read_mmCIF_file(str(cif_file)).get_contacts(test_CA, threshold, auth_chain1, auth_chain2, auth_contacts=True, auth_chain_id_supplied=True)}
+        fetch_pdb_contacts = {(int(x[0]), int(x[1])) for x in PDBInformation.fetch_pdb(pdb_id, struc_format="pdb").get_contacts(test_CA, threshold, auth_chain1, auth_chain2, auth_contacts=True)}
+        read_pdb_contacts = {(int(x[0]), int(x[1])) for x in PDBInformation.read_pdb_file(f"tests/pdb_info/{pdb_id.lower()}.pdb").get_contacts(test_CA, threshold, auth_chain1, auth_chain2, auth_contacts=True)}
+        
+        
+        fetch_cif_contacts = drop_inord_res(fetch_cif_contacts)
+        fetch_authchain_cif_contacts = drop_inord_res(fetch_authchain_cif_contacts)
+        read_cif_contacts = drop_inord_res(read_cif_contacts)
+        read_authchain_cif_contacts = drop_inord_res(read_authchain_cif_contacts)
+        fetch_pdb_contacts = drop_inord_res(fetch_pdb_contacts)
+        read_pdb_contacts = drop_inord_res(read_pdb_contacts)
+        
+        if pdbx.get_model_count(pdbx.CIFFile.read(rcsb.fetch(pdb_id, format="mmcif"))) <= 1:
+            compare(fetch_cif_contacts, fetch_pdb_contacts, read_cif_contacts, read_pdb_contacts)
+            print(f"{pdb_id} has no difference between cif and pdb reading.")
+
+            compare(fetch_cif_contacts, fetch_authchain_cif_contacts, read_cif_contacts, read_authchain_cif_contacts)
+            print(f"{pdb_id} has no issue reading with auth chains and asym chains.")
+
+            compare(fetch_cif_contacts, cif_file_contacts)
+            print(f"{pdb_id} has no difference between cif and interface_contacts_X.py reading.")
+        else:
+            compare(fetch_cif_contacts, fetch_pdb_contacts, read_cif_contacts, read_pdb_contacts)
+            print(f"{pdb_id} has no difference between cif and pdb reading.")
+
+            compare(fetch_cif_contacts, fetch_authchain_cif_contacts, read_cif_contacts, read_authchain_cif_contacts)
+            print(f"{pdb_id} has no issue reading with auth chains and asym chains.")
+        
+def test_contacts():
+    check_contacts(test_CA=True, threshold=10)
+    check_contacts(test_CA=False, threshold=8)

dcatoolkit-0.1.8/src/dcatoolkit/__init__.py

@@ -1,4 +0,0 @@
-
-__version__ = "0.1.8"
-from .representation import Pairs, DirectInformationData, StructureInformation, ResidueAlignment
-from .analytics import MSATools