modified code (__init__) to save distance matrixes in Conformations instances

src/projet8.py

 
     def __init__(self, filename):
         """
-        df : pd.DataFrame object
+        filename : a .pdb file or several pdb file containing the conformations
+
+        Attribute :
+            - df : pd.DataFrame object
+            - filename : a list of str, paths to the structure files
+            - simple_dist : matrix distance computed with the identity method
+                            Initiality empty, must be computed
+            - dissimilarity_dist : matrix distance computed with the
+                                   dissimilarity method, initialy empty because
+                                   the computation time is long
         Each row of the dataframe is a conformation and each column a position
         of the sequence.
         """
         self.df = pd.DataFrame()
+        self.filename = filename
+        self.identity_dist = None
+        self.dissimilarity_dist = None
         for chain_name, chain in pbx.chains_from_files([filename]): 
                 dihedrals = chain.get_phi_psi_angles() 
                 pb_seq = pbx.assign(dihedrals) 
  
     def identity(self):
         """
-        Returns a matrix of the distance between all conformations based on
+        Computes a distance matrix between all conformations based on
         wether the PB at each position is identical or not.
         Returns the matrix as a numpy.ndarray object.
         """
                              'o':15,'p':16} # dict to transform PB to int,
                                             # necessary for the pdsit function
         dfnum = self.df.replace(dict_str_to_float)
+        dist = squareform(pdist(dfnum, metric ='jaccard'))
+        self.identity_dist = dist
 
-        return squareform(pdist(dfnum, metric ='jaccard'))
+        return dist
 
     def dissimilarity(self, matrix=None):
         """
                     dissimilarity[i][j] += matrix[confs.df.loc[i,k]][confs.df.loc[j,k]]
         dissimilarity = dissimilarity + dissimilarity.T # fills the whole matrix
         
-        return dissimilarity
+        return self.dist_from_dissimilarity(dissimilarity)
 
-    def dist_from_dissimilarity(diss_matrix):
+    def dist_from_dissimilarity(self, diss_matrix):
         """
         Using the substitution matrix from the PBxplore package, the obtained
         dissimilarity matrix has both positive and negative values. Low value
         diss_matrix = -diss_matrix
         diss_matrix = (diss_matrix - np.min(diss_matrix))/np.ptp(diss_matrix)
         dist = diss_matrix * abs((np.identity(confs.df.shape[0])-1))
-        
+
+        self.dissimilarity_dist = dist
+
         return dist
+    
+    #def small_kmedoid(self, ncluster):
+    #    pass
+    
 
 if __name__ == "__main__":
-    if len(sys.argv) != 2:
+    if len(sys.argv) < 2:
         sys.exit("Error : usage '$ python3 projet8 md.pdb'")
-
+    
     confs = Conformations(sys.argv[1])
-
+    
 
     print(confs.df)