Blunt sidechains for mk_prepare_receptor?

[rwxayheee]

Hi all,

Would it be an appealing change to you, if residues with partially resolved are modeled as Ala by mk_prepare_receptor?

In many crystal structures, methyl groups are used as placeholders for unresolved sidechains. Some examples can be found in PDB ID 1FPU (receptor in the Vina flexible docking example). They all have backbones and CB:

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     ASP A 226    CG   OD1  OD2                                       
REMARK 470     SER A 229    OG                                                  
REMARK 470     PRO A 230    CG   CD                                             
REMARK 470     ASN A 231    CG   OD1  ND2                                       
REMARK 470     LYS A 274    CG   CD   CE   NZ                                   
REMARK 470     GLU A 275    CG   CD   OE1  OE2                                  
REMARK 470     ASP A 276    CG   OD1  OD2                                       
REMARK 470     THR A 277    OG1  CG2                                            
REMARK 470     MET A 278    CG   SD   CE                                        
REMARK 470     GLU A 279    CG   CD   OE1  OE2                                  
REMARK 470     LYS A 285    CG   CD   CE   NZ                                   
REMARK 470     LYS A 294    CG   CD   CE   NZ                                   
REMARK 470     THR A 306    OG1  CG2                                            
REMARK 470     ARG A 307    CG   CD   NE   CZ   NH1  NH2                        

Modeling them as Ala is like placing blunt sidechains on them. The two major advantages are:
Preserving the CB might be helpful for further repairments of the structure.
Not deleting them with --allow_bad_res avoids unnecessary segmentation of the backbone.

What do you think? The feature is not of very high priority. Happy to submit a PR if anyone is interested

[diogomart]

I think it's a good idea as long as all backbone atoms and C-beta are present. For charged residues, we could further consider preserving the net charge with some trick as discussed with @mattholc