DR_bind DNA-binding residues prediction failure diagnosing page

If your DR_bind has failed for some reason you should get a webpage that looks like this below:

Hi there, it seems that your DNA-binding residue prediction on 1BGW chain A failed for some reason.

Status: Job FAILED

Our experience is that this is normally related to problems with the PDB file, often duplicated residues or residues that are missing backbone atoms. If you can please take the 'cleaned' PDB file and check it over and resubmit it via option A.

Often the problem residue is listed at the end of the last error file below. Also you can email (removed for this example) with this url so we can investigate the problem, Thanks.

The failed script was run1.pl so the error file to look at first should be run1.err.

The cleaned pdb file we're working with was: 1BGW.pdb
The original pdb file was: 1BGW.orig
The ConSurf file was: 1BGW.consurf

Log files:
leap.log,

Error files:
run0.err,
run1.err,

Diagnosing the problem

Our experience is that the problem usually lies with the PDB file, however the first place to look is at the final few lines of the last error message. When DR_bind runs the scripts are in sequence (0-4) so the highest numbered error file is the one to start with. If we look at run1.err and the final few lines we see this:
Final few lines of the run1.err file.

cluster 7
Number of triangles 35, average gap depth =    7.218

Cluster 8
Number of triangles 9, average gap depth =    7.668

Cluster 9
Number of triangles 14, average gap depth =    6.326
2 mark cleft
1077 Structure  Sequence ALA

Which seems to indicate a problem with the Ala1077 reisdue. If we inspect the 1BGW.pdb file which is the 'cleaned' version of the original and look for the Ala1077 reisdue we see this:
Checking 'cleaned' 1BGW.pdb for Ala1077 residue.

ATOM   6073  CB  ILE A1076     122.681 156.401  59.158  1.00137.50        
ATOM   6074  H   ILE A1076     121.598 158.903  60.075  1.00  0.00
ATOM   6075  N   ALA A1077     123.158 158.258  62.279  1.00131.77 >      
ATOM   6076  N   GLU A1108     123.717 151.061  59.017  1.00120.02  <     
ATOM   6077  CA  GLU A1108     122.601 150.156  58.617  1.00120.05  <     
ATOM   6078  C   GLU A1108     123.185 149.162  57.660  1.00119.32  <     
ATOM   6079  O   GLU A1108     124.400 149.076  57.492  1.00119.90  <

The obvious thing here is that there is only one atom listed for the Ala1077, the backbone N atom. Checking the original PDB file (1BGW.orig) for this residue:
Checking original 1BGW.orig for Ala1077 residue.

ATOM   6073  CB  ILE A1076     122.681 156.401  59.158  1.00137.50           C  
ATOM   6074  H   ILE A1076     121.598 158.903  60.075  1.00  0.00           H  
ATOM   6075  N   ALA A1077     123.158 158.258  62.279  1.00131.77           N  
ATOM   6076  N   GLU A1108     123.717 151.061  59.017  1.00120.02           N  
ATOM   6077  CA  GLU A1108     122.601 150.156  58.617  1.00120.05           C  
ATOM   6078  C   GLU A1108     123.185 149.162  57.660  1.00119.32           C  
ATOM   6079  O   GLU A1108     124.400 149.076  57.492  1.00119.90           O

We can see that the backbone atoms are not in the orignal PDB file either, infact in the MISSING record for 1BGW we see that there are known to be missing:
Missing atoms lines in original 1BGW.orig file.

REMARK 470 MISSING ATOM                                                         
REMARK 470 THE FOLLOWING RESIDUES HAVE MISSING ATOMS(M=MODEL NUMBER;            
REMARK 470 RES=RESIDUE NAME; C=CHAIN IDENTIFIER; SSEQ=SEQUENCE NUMBER;          
REMARK 470 I=INSERTION CODE):                                                   
REMARK 470   M RES CSSEQI  ATOMS                                                
REMARK 470     SER A1071    OG                                                  
REMARK 470     PRO A1072    CG   CD                                             
REMARK 470     ASN A1073    CG   OD1  ND2                                       
REMARK 470     ASP A1074    CG   OD1  OD2                                       
REMARK 470     GLU A1075    CG   CD   OE1  OE2                                  
REMARK 470     ILE A1076    CG1  CG2  CD1                                       
REMARK 470     ALA A1077    CA   C    O    CB

However if we check the 1BGW.ConSurf file we see that the ConSurf file includes the Ala1077:
Checking the 1BGW.consurf file for the Ala1077 residue.

 667       I     ILE1076:A       1.986            1      2.250, 2.250                       1,1                    17/74        Q,T,
D,I,K,G,E,L,V
 668       A     ALA1077:A       1.534            1      0.969, 2.250                       2,1                    16/74        A,N,K,E,V,Q,D,I,L
 669       E     GLU1108:A       0.708            3      0.476, 0.969                       3,2                    74/74        S,A,F,N,P,E,Y,Q,D,C,I,L

We see that the Ala1077 residue is included in the ConSurf file, the most likely cause of the problem is that the missing backbone atoms are creating an issue.

The solution.

There are at least two solutions to this problem,
a) use a molecular editor to add in the missing C, Ca and O atoms to the PDB file and resubmit via option A,
b) to remove the Ala1077 N atom and the 1077 residue from the ConSurf file and resubmit via option A.

We did indeed remove the Ala1077 N atom and the prediction continued.

DR_bind is hosted at The Institute of Biomedical Sciences, Academia Sinica, Taipei 11529, Taiwan.