[phenixbb] pdb help

Pavel Afonine pafonine at lbl.gov
Fri May 19 12:35:34 PDT 2017


Hi,

I afraid for very large molecules it may take a lot of memory. This is 
because internally the procedure generates and accumulates hundreds if 
not thousands maps.. You can try to make these maps smaller by changing
resolution_factor=1./4
to
resolution_factor=1./3

which may diminish the efficacy of the protocol but that's the only 
suggestion I have (apart from trying it on a "bigger" computer).

Pavel

On 5/18/17 14:28, jp d wrote:
> hi,
>
> phenix version 1.11.1-2575
>
> FEM has crashed a few times for me .
> i have tried using various parameters (resolution_factor=1./3, use_resolve=False) ,
> but it will max out my RAM (24G) and get into swap space.
> It gets very memory intensive, i think thats why it crashed.
>
> Do you have any suggestions to try ?
>
> i previously had a similar problem with a pdb using conflicting chain assignments,
> but this pdb doesn't use chains
>
> thanks
> jpd
> _______________________________________________
> phenixbb mailing list
> phenixbb at phenix-online.org
> http://phenix-online.org/mailman/listinfo/phenixbb
> Unsubscribe: phenixbb-leave at phenix-online.org



More information about the phenixbb mailing list