Merge pull request #37 from Pranavkhade/get_sequence

get_sequence added.

packman/__init__.py

@@ -15,4 +15,4 @@
 
 
 #VERSION CHANGE HERE CHANGES IT IN docs AND setup.py
-__version__='1.4.6'
+__version__='1.4.7'

packman/constants/Constants.py

@@ -112,6 +112,34 @@
 'U':238.02891,
 }
 
+#Kidera factors
+kidera_factors={
+'ALA':[-1.56,-1.67,-0.97,-0.27,-0.93,-0.78,-0.2,-0.08,0.21,-0.48],
+'ARG':[0.22,1.27,1.37,1.87,-1.7,0.46,0.92,-0.39,0.23,0.93],
+'ASN':[1.14,-0.07,-0.12,0.81,0.18,0.37,-0.09,1.23,1.1,-1.73],
+'ASP':[0.58,-0.22,-1.58,0.81,-0.92,0.15,-1.52,0.47,0.76,0.7],
+'CYS':[0.12,-0.89,0.45,-1.05,-0.71,2.41,1.52,-0.69,1.13,1.1],
+'GLN':[-0.47,0.24,0.07,1.1,1.1,0.59,0.84,-0.71,-0.03,-2.33],
+'GLU':[-1.45,0.19,-1.61,1.17,-1.31,0.4,0.04,0.38,-0.35,-0.12],
+'GLY':[1.46,-1.96,-0.23,-0.16,0.1,-0.11,1.32,2.36,-1.66,0.46],
+'HIS':[-0.41,0.52,-0.28,0.28,1.61,1.01,-1.85,0.47,1.13,1.63],
+'ILE':[-0.73,-0.16,1.79,-0.77,-0.54,0.03,-0.83,0.51,0.66,-1.78],
+'LEU':[-1.04,0,-0.24,-1.1,-0.55,-2.05,0.96,-0.76,0.45,0.93],
+'LYS':[-0.34,0.82,-0.23,1.7,1.54,-1.62,1.15,-0.08,-0.48,0.6],
+'MET':[-1.4,0.18,-0.42,-0.73,2,1.52,0.26,0.11,-1.27,0.27],
+'PHE':[-0.21,0.98,-0.36,-1.43,0.22,-0.81,0.67,1.1,1.71,-0.44],
+'PRO':[2.06,-0.33,-1.15,-0.75,0.88,-0.45,0.3,-2.3,0.74,-0.28],
+'SER':[0.81,-1.08,0.16,0.42,-0.21,-0.43,-1.89,-1.15,-0.97,-0.23],
+'THR':[0.26,-0.7,1.21,0.63,-0.1,0.21,0.24,-1.15,-0.56,0.19],
+'TRP':[0.3,2.1,-0.72,-1.57,-1.16,0.57,-0.48,-0.4,-2.3,-0.6],
+'TYR':[1.38,1.48,0.8,-0.56,0,-0.68,-0.31,1.03,-0.05,0.53],
+'VAL':[-0.74,-0.71,2.04,-0.4,0.5,-0.81,-1.07,0.06,-0.46,0.65],
+'ASX':[0.86,-0.145,-0.85,0.81,-0.37,0.26,-0.805,0.85,0.93,-0.515],
+'GLX':[-0.96,0.215,-0.77,1.135,-0.105,0.495,0.44,-0.165,-0.19,-1.225],
+'XAA':[0,0,0,0,0,0,0,0,0,0],
+'XLE':[-0.885,-0.08,0.775,-0.935,-0.545,-1.01,0.065,-0.125,0.555,-0.425]
+}
+
 
 #van der Waals Volumes and Radii, A. Bondi, The Journal of Physical Chemistry 1964 68 (3), 441-451, DOI: 10.1021/j100785a001
 vdw_surface_bondi={
@@ -234,3 +262,37 @@
 'Ts':None,
 'Og':None
 }
+
+#Amino acid three letter code to one letter code
+THREE_LETTER_TO_ONE_LETTER = {
+'ALA':'A',
+'ARG':'R',
+'ASN':'N',
+'ASP':'D',
+'CYS':'C',
+'GLN':'Q',
+'GLU':'E',
+'GLY':'G',
+'HIS':'H',
+'ILE':'I',
+'LEU':'L',
+'LYS':'K',
+'MET':'M',
+'PHE':'F',
+'PRO':'P',
+'PYL':'O',
+'SER':'S',
+'SEC':'U',
+'THR':'T',
+'TRP':'W',
+'TYR':'Y',
+'VAL':'V',
+'ASX':'B',
+'GLX':'Z',
+'XAA':'X',
+'XLE':'J',
+'UNK':'X'
+}
+
+#Amino acid one letter code to three letter code
+ONE_LETTER_TO_THREE_LETTER = {THREE_LETTER_TO_ONE_LETTER[three_letter]:three_letter for three_letter in THREE_LETTER_TO_ONE_LETTER}

packman/constants/__init__.py

@@ -1,3 +1,6 @@
 from .Constants import amino_acid_molecular_weight
 from .Constants import atomic_weight
-from .Constants import vdw_surface_bondi
+from .Constants import vdw_surface_bondi
+from .Constants import kidera_factors
+from .Constants import THREE_LETTER_TO_ONE_LETTER
+from .Constants import ONE_LETTER_TO_THREE_LETTER

packman/molecule/chain.py

@@ -309,6 +309,17 @@ def get_backbone(self):
         """
         return [i.get_backbone() for i in self.get_residues()]
 
+    def get_sequence(self):
+        """Get the Amino acid sequence of the chain. (Protein chains only)
+        
+        Returns:
+            FASTA format string of the chain sequence.
+        """
+        try:
+            return '>'+str(self.get_parent().get_parent().get_id())+'_'+str(self.get_parent().get_id())+'_'+str(self.get_id())+'\n'+''.join([resi.get_changed_alphabet() if len(resi.get_name())>1 else resi.get_name() for resi in self.get_residues()])
+        except:
+            logging.debug('Error in retriving chain sequence.')
+
     def calculate_entropy(self,entropy_type,chains=None, probe_size=1.4, onspherepoints=30):
         """Calculate the entropy for the each amino acid will be returned.
     

packman/molecule/model.py

@@ -90,7 +90,10 @@ def __init__(self,id,AllAtoms,AllResidues,AllChains,AllHetAtoms,AllHetMols):
         self.__properties = {}
 
     def __getitem__(self,ChainID):
-        return self.__AllChains[ChainID]
+        try:
+            return self.__AllChains[ChainID]
+        except KeyError:
+            logging.warning('Please provide a valid chain ID.')
 
     #Get Functions
     def get_id(self):
@@ -348,6 +351,14 @@ def get_torsion(self, bond, neighbor1=None, neighbor2=None, radians=True):
         else:
             return numpy.rad2deg(radang)
 
+    def get_sequence(self):
+        """Get the Amino acid sequence of the chain. (Protein chains only)
+        
+        Returns:
+            FASTA format string of the chain sequence.
+        """
+        return '\n'.join( [chain.get_sequence() for chain in self.get_chains() if chain.get_sequence()!=''] )
+
         #Set Functions
     def set_id(self,new_id):
         """Set the ID of the given 'Model'
@@ -667,7 +678,7 @@ def calculate_bonds(self):
                     N.set_bond(bond)
                     self.__ModelGraph.add_edge( C.get_id(), N.get_id() , id = counter )
                 else:
-                    logging.warning('The peptide bond following residue is missing: '+str(resi[i].get_id())+' Chain: '+resi[i].get_parent().get_id())
+                    logging.info('The peptide bond following residue is missing: '+str(resi[i].get_id())+' Chain: '+resi[i].get_parent().get_id())
 
         try:
             None

packman/molecule/molecule.py

@@ -136,7 +136,7 @@ def load_pdb(filename):
 
     if(len(Models)>2):
         #NMR
-        logging.warning('Multiple models/frames are detected (B-factor field is now a calculated parameter, i.e., the scalar standard deviation of the atom location of all frames)')
+        logging.debug('Multiple models/frames are detected (B-factor field is now a calculated parameter, i.e., the scalar standard deviation of the atom location of all frames)')
         All_Coords=[]
         for i in Models:
             All_Coords.append(numpy.array([j.get_location() for j in i.get_atoms()]))
@@ -415,7 +415,7 @@ def load_cif(filename):
 
     if(len(AllModels)>2):
         #NMR
-        logging.warning('Multiple models/frames are detected (B-factor field is now a calculated parameter, i.e., the scalar standard deviation of the atom location of all frames)')
+        logging.info('Multiple models/frames are detected (B-factor field is now a calculated parameter, i.e., the scalar standard deviation of the atom location of all frames)')
         All_Coords=[]
         for i in AllModels:
             All_Coords.append(numpy.array([j.get_location() for j in i.get_atoms()]))

packman/molecule/protein.py

@@ -25,6 +25,7 @@
 """
 
 from numpy import around
+from . import model
 
 class Protein():
     """This class contains the information about the 'Protein' object (packman.molecule.Protein).
@@ -60,6 +61,14 @@ def get_id(self):
         """
         return self.__id
 
+    def get_models(self):
+        """Get all the models at once. Useful for the iterations.
+
+        Returns:
+            Generator of all the models in the PDB file.
+        """
+        yield self.__Models
+
     def get_data(self):
         """Get the misc data (other than coordinates) from the file.
 
@@ -68,6 +77,23 @@ def get_data(self):
         """
         return self.__Data
 
+    def get_sequence(self, all_models = False):
+        """_summary_
+
+        Args:
+            all_models (bool, optional): Get sequence of the all frames; useless if sequence accross the models is identical. Defaults to False.
+
+        Returns:
+            Protein sequence in FASTA format.
+        """
+        for model in self:
+            try:
+                return model.get_sequence()
+            except:
+                None
+            if(all_models==False):
+                break
+
     #Set functions
     def set_data(self,data):
         """Set the misc data (other than coordiantes) to the Protein object.

packman/molecule/residue.py

@@ -25,6 +25,8 @@
 import numpy
 import logging
 
+from ..utilities import change_alphabet
+
 
 class Residue():
     """This class contains the information about the 'Residue' object (packman.molecule.Residue).
@@ -165,6 +167,14 @@ def get_property(self,property_name):
             return self.__properties[property_name]
         except:
             logging.warning('The Property Name provided is not assigned.')
+
+    def get_changed_alphabet(self):
+        """Converts three letter amino acid code to one letter and vise-versa
+
+        Returns:
+            AA (string) : Three or one letter amino acid code depending and opposite of the argument provided.
+        """
+        return change_alphabet(self.get_name())
 
     #Set Functions
     def set_id(self,new_id):

packman/tests/molecule/test_molecule.py

@@ -29,6 +29,7 @@ def test_Protein(self):
         rm( 'test.pdb' )
 
         self.assertNotEqual( len([i for i in self.mol]) , 0 )
+        self.assertNotEqual( len( [i for i in self.mol.get_sequence()] ), 0 )
 
     def test_Model(self):
         #Basic
@@ -42,6 +43,7 @@ def test_Model(self):
         self.assertNotEqual( len( [i for i in self.mol[0].get_atoms()] )   , 0 )
         self.assertNotEqual( len( [i for i in self.mol[0].get_calpha()] )  , 0 )
         self.assertNotEqual( len( [i for i in self.mol[0].get_backbone()] ), 0 )
+        self.assertNotEqual( len( [i for i in self.mol[0].get_sequence()] ), 0 )
 
     def test_Chain(self):
         #Basic
@@ -54,6 +56,7 @@ def test_Chain(self):
         self.assertNotEqual( len( [i for i in self.mol[0].get_atoms()] )   , 0 )
         self.assertNotEqual( len( [i for i in self.mol[0].get_calpha()] )  , 0 )
         self.assertNotEqual( len( [i for i in self.mol[0].get_backbone()] ), 0 )
+        self.assertNotEqual( len( [i for i in self.mol[0]['A'].get_sequence()] ), 0 )
 
     def test_Residue(self):
         #Basic

packman/utilities/utilities.py

@@ -189,16 +189,18 @@ def change_alphabet(AA):
     Returns:
         AA (string) : Three or one letter amino acid code depending and opposite of the argument provided.
     """
-    three_to_one_lookup = { 'ALA': 'A', 'ARG': 'R', 'ASN': 'N', 'ASP': 'D', 'CYS': 'C', 'GLN': 'Q', 'GLU': 'E', 'GLY': 'G', 'HIS': 'H', 'ILE': 'I', 'LEU': 'L', 'LYS': 'K', 'MET': 'M', 'PHE': 'F', 'PRO': 'P', 'SER': 'S', 'THR': 'T', 'TRP': 'W', 'TYR': 'Y', 'VAL': 'V' }
-    one_to_three_lookup = {three_to_one_lookup[i]:i for i in three_to_one_lookup.keys()}
+    from ..constants import Constants
+
+    three_to_one_lookup = Constants.THREE_LETTER_TO_ONE_LETTER
+    one_to_three_lookup = Constants.ONE_LETTER_TO_THREE_LETTER
 
     try:
         return three_to_one_lookup[AA]
     except:
         try:
             return one_to_three_lookup[AA]
         except:
-            logging.warning('Amino acid code provided did not match any of three or one letter code')
+            logging.warning('Amino acid code provided did not match any of three or one letter code; returning unknown amino acid code.')
             if(len(AA)==3):
                 return 'X'
             if(len(AA)==1):