ADL: dont know what this means...???

Garrett M. Morris garrett at scripps.edu
Sun Mar 30 16:43:08 PDT 2008


Dear Namitha,

Here is a quick update: the warning message refers to discrepancies,  
and the version of ADT you have appears to have a more stringent  
tolerance (later versions use 0.07).  Read on for details:

On Mar 28, 2008, at 12:06 pm, Garrett M. Morris wrote:

>
> Dear Namitha,
>
> On 28 Mar, 2008, at 4:27 am, namita at ncbs.res.in wrote:
>
>> Dear users
>>
>> Thanks for all the help given so far in helping me solve my issues.
>> To an
>> extent I have succeeded in resolving many of my problems but again I
>> face
>> many more...pls do help me this time again...
>>
>> I try to save in my receptor molecule (which is generated by
>> modeller) in
>> pdbqt format. (Grid -> Macromolecule -> Choose)and then saves the
>> molecule
>> in pdbqt. when I try opening this molecule again it asks me whether
>> i need
>> to preserve charges or add gasteiger charges to it. Either i say no
>> or yes
>> to that option it gives me a warning message like this :
>>
>> Non-integral charge on molecule:- -6.075
>>
>> correct 11 residues:
>> HIS1 0.055
>> GLN13 0.006
>> LYS27 0.006
>> TYR44 0.03
>> PRO45 0.03
>> GLU69 0.029
>> PRO70 0.03
>> ARG104 0.029
>> PRO105 0.03
>> ASN196 0.006
>> PRO473 0.058
>> Charges should be corrected in written output file!

NOTE: These numbers are the _discrepancies_ from the integer values,  
not the absolute total charges.

>>
>> I would like to know how and where do i need to add these charges?  
>> Are
>> they talking about the .pdbqt output file??
>
> Yes.
>
>>
>>
>> Please help...any help would be greatly appreciated.
>
> Sorry if this message from ADT is not clear.
>
> One check ADT does on the macromolecule is to add up all the partial
> charges on each residue, to see if they sum to an integer.  The idea
> is that a side chain can be neutral or charged, but never carry a
> fractional electron...  This kind of error check can detect when there
> are missing atoms in a sidechain, something that is important in
> guaranteeing your molecular model is accurate.
>
> These discrepancies are very small for each amino acid, so I would not
> worry about them too much.  I will talk with the ADT programmers about
> this.

I spoke to Ruth Huey, and she informed me that:

> The tolerance was increased 2/23/07 to +/-0.07 in AutoDockTools/ 
> autotorsCommands.py.
>
> This last release which had .005, in principle, was 1.4.4 dated  
> 5/24/2007
> Releases dated 10/03/07 and later have the larger tolerance.


I would recommend you obtain the most recent version of ADT: is has  
many new features as well as bug-fixes:

http://autodock.scripps.edu/resources/adt/index_html

Version 1.5.0 is the latest "non-beta" version:

http://mgltools.scripps.edu/downloads/previous-releases/mgltools-1-5.0

>   You can actually edit the charges in the PDBQT file using a
> text editor.
>
> Thanks,
>
> Garrett
>
>>
>>
>> Thanks
>> Namitha
>>
>> ________________________________________________
>> --- ADL: AutoDock List  --- http://autodock.scripps.edu/mailing_list
>> ---








More information about the autodock mailing list