Source Code for Module Bio.FSSP.FSSPTools

 1  from Bio import FSSP 
 2  import copy 
 3  from Bio.Align import Generic 
 4  from Bio import Alphabet 
 5   
 6   
 7 -class FSSPAlign(Generic.Alignment): 
 8 -    def _add_numbering_table(self, new_record): 
 9          new_record.annotations['abs2pdb'] = {} 
10          new_record.annotations['pdb2abs'] = {} 
11   
12   
13 -class FSSPMultAlign(dict): 
14 -    def __init__(self): 
15          self.abs_res = [] 
16          self.pdb_res = [] 
17          self.data = {} 
18   
19   
20 -def mult_align(sum_dict, align_dict): 
21      """Returns a biopython multiple alignment instance (Bio.Align.Generic)""" 
22      mult_align_dict = {} 
23      for j in align_dict.abs(1).pos_align_dict: 
24          mult_align_dict[j] = '' 
25   
26      for i in range(1, len(align_dict)+1): 
27          # loop on positions 
28          for j in align_dict.abs(i).pos_align_dict: 
29              # loop within a position 
30              mult_align_dict[j] += align_dict.abs(i).pos_align_dict[j].aa 
31      seq_order = mult_align_dict.keys() 
32      seq_order.sort() 
33      fssp_align = Generic.Alignment(Alphabet.Gapped( 
34                                     Alphabet.IUPAC.extended_protein)) 
35      for i in seq_order: 
36          fssp_align.add_sequence(sum_dict[i].pdb2+sum_dict[i].chain2, 
37                                  mult_align_dict[i]) 
38  #        fssp_align._add_numbering_table() 
39      return fssp_align 
40   
41   
42  # Several routines used to extract information from FSSP sections 
43  # filter: 
44  # filters a passed summary section and alignment section according to a numeric 
45  # attribute in the summary section. Returns new summary and alignment sections 
46  # For example, to filter in only  those records which have a zscore greater than 
47  # 4.0 and lesser than 7.5: 
48  # new_sum, new_align = filter(sum, align, 'zscore', 4, 7.5) 
49  # 
50  # Warning: this function really slows down when filtering large FSSP files. 
51  # The reason is the use of copy.deepcopy() to copy align_dict into 
52  # new_align_dict. I have to figure out something better. 
53  # Took me ~160 seconds for the largest FSSP file (1reqA.fssp) 
54  # 
55   
56 -def filter(sum_dict, align_dict, filter_attribute, low_bound, high_bound): 
57      """filters a passed summary section and alignment section according to a numeric 
58      attribute in the summary section. Returns new summary and alignment sections""" 
59      new_sum_dict = FSSP.FSSPSumDict() 
60      new_align_dict = copy.deepcopy(align_dict) 
61  #    for i in align_dict: 
62  #        new_align_dict[i]  = copy.copy(align_dict[i]) 
63     # new_align_dict = copy.copy(align_dict) 
64      for prot_num in sum_dict: 
65          attr_value = getattr(sum_dict[prot_num], filter_attribute) 
66          if attr_value >= low_bound and attr_value <= high_bound: 
67              new_sum_dict[prot_num] = sum_dict[prot_num] 
68      prot_numbers = new_sum_dict.keys() 
69      prot_numbers.sort() 
70      for pos_num in new_align_dict.abs_res_dict: 
71          new_align_dict.abs(pos_num).pos_align_dict = {} 
72          for prot_num in prot_numbers: 
73              new_align_dict.abs(pos_num).pos_align_dict[prot_num] = \ 
74                      align_dict.abs(pos_num).pos_align_dict[prot_num] 
75      return new_sum_dict, new_align_dict 
76   
77   
78 -def name_filter(sum_dict, align_dict, name_list): 
79      """ Accepts a list of names. Returns a new Summary block and Alignment block which 
80          contain the info only for those names passed.""" 
81      new_sum_dict = FSSP.FSSPSumDict() 
82      new_align_dict = copy.deepcopy(align_dict) 
83      for cur_pdb_name in name_list: 
84          for prot_num in sum_dict: 
85              if sum_dict[prot_num].pdb2+sum_dict[prot_num].chain2 == cur_pdb_name: 
86                  new_sum_dict[prot_num] = sum_dict[prot_num] 
87      prot_numbers = new_sum_dict.keys() 
88      prot_numbers.sort() 
89      for pos_num in new_align_dict.abs_res_dict: 
90          new_align_dict.abs(pos_num).pos_align_dict = {} 
91          for prot_num in prot_numbers: 
92              new_align_dict.abs(pos_num).pos_align_dict[prot_num] = \ 
93                      align_dict.abs(pos_num).pos_align_dict[prot_num] 
94      return new_sum_dict, new_align_dict 
95