[phenixbb] phenix.automr high zscore but too many clashes

[Andreas Förster]

Hey Zach,

remove the loops from your search model and rerun the MR. 
Alternatively, increase the maximum number of clashes allowed until you 
get a solution written out.  See where your solutions clash.  Again, if 
it's in a loop region, cut the loop and rerun.  Your z-scores look 
great.  (Notice that you have zed-scores when you run phaser through 
ccp4i and zee-scores when you run phaser through phenix.)


Andreas



zach powers wrote:
> Hi,
> 
> 
> I have a problem and i wonder if someone has had a similar experience. I 
> have been using phenix.automr program to find an MR solution for  a 
> protein.  When I ask Phaser to use copies=2 I find a number of solutions 
> with poor z-scores (3-4). When I ask it to use copies=3, I get great 
> Z-scores (see below) but no solution due to the high number of clashes 
> (>100!).
> 
> As an x-ray newbie I am a bit perplexed of what to make of this: the 
> solution is not good because there is not enough space in the unit cell 
> to comfortably accommodate all three molecules yet the Z-score indicates 
> the structure is good.
> 
> My structure does have several loops and these may be contributing to 
> the clashing residues. As a newbie I have a newbie question - what does 
> this mean (great z-score but no solutions)? Is this non-solution a 
> possible solution if I play with it, or is the packing simply too tight 
> and the Z-scores are not valid?
> 
> In the meantime I am chopping my protein into sub-domains as recommended 
> in the Phaser documentation, but if anyone has seen something like this 
> before, any suggestions are welcome.
> 
> thanks
> zach charlop-powers
> 
> 
> 
> 
> Packing Table: Space Group P 3
>    ------------------------------
>    Solutions accepted if number of clashes = 135 (lowest number of 
> clashes in list)
>    provided this number of clashes <= 10 (maximum number of allowed clashes)
>    #  #Clashes #   Accepted Annotation
>    1   188         NO       RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 
> TFZ=10.4
>    2   156         NO       RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 
> TFZ=10.2
>    3   151         NO       RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 
> TFZ=9.8
>    4   146         NO       RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 
> TFZ=9.7
>    5   187         NO       RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=16.3 
> TFZ=10.5
>    6   202         NO       RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 
> TFZ=9.2
>    7   170         NO       RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 
> TFZ=9.1
>    8   144         NO       RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=16.3 
> TFZ=10.2
>    9   177         NO       RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 
> TFZ=9.1
>    10  153         NO       RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 
> TFZ=8.9
>    11  150         NO       RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 
> TFZ=8.8
>    12  217         NO       RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 
> TFZ=8.6
>    13  182         NO       RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 
> TFZ=8.6
>    14  135         NO       RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 
> TFZ=8.5
>    15  175         NO       RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 
> TFZ=8.3
>    16  159         NO       RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 
> TFZ=8.3
>    17  192         NO       RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 
> TFZ=8.3
>    18  165         NO       RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=16.3 
> TFZ=9.2
>    19  169         NO       RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=16.3 
> TFZ=9.1
>    20  158         NO       RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 
> TFZ=8.0
>    21  173         NO       RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 
> TFZ=8.0
>    22  154         NO       RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=16.3 
> TFZ=8.9
>    23  206         NO       RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 
> TFZ=7.8
> 
>    0 accepted of 23 solutions
> _______________________________________________
> phenixbb mailing list
> phenixbb at phenix-online.org
> http://www.phenix-online.org/mailman/listinfo/phenixbb
> 

-- 
         Andreas Förster, Research Associate
         Paul Freemont & Xiaodong Zhang Labs
Department of Biochemistry, Imperial College London
             http://www.msf.bio.ic.ac.uk