[phenixbb] mrage with nucleic acids?
gb360 at cam.ac.uk
Mon Nov 17 03:26:34 PST 2014
unfortunately, mrage does not currently handle nucleic acids.
However, I am wondering whether it would be beneficial to add this
feature. My understanding is that for nucleic acids, the
sequence-structure relation is not the same as for proteins, therefore
the majority of MRage's features, such as homology search and template
editing are not applicable, but I am glad to listen to the experts on
the field, and add features that would be make the program more useful
(with this said, I am currently adding support for sculptor to process
nucleic acid chains, although right now this is not more than renaming
to the expected sequence; any advice on further possibilities is
If you just have one model, and want to run it through MRage, you can do
this by leaving out the component sequence, and specify the model
structure as a "processed model" (I am not sure that the GUI lets you do
this, but you can do it from the command line); in this case, the
molecular weight is calculated from the structure (and a warning will be
printed to the logfile). This will still assume that the structure is
protein, and hence estimate SigmaN incorrectly, but the error is rather
small. However, unless you have a large number of pre-made models and
want to run it on a cluster, this is unlikely to give better results
than running Phaser to do the same.
Best wishes, Gabor
On 2014-11-16 15:17, Paul Paukstelis wrote:
> I'm trying to run mrage with nucleic acid models. With phaser, it is
> possible to specify the molecule type, but this option doesn't exist
> in mrage (from what I can see). Providing a sequence file
> automatically assumes protein one letter codes. Any suggestion on how
> to clearly define a sequence file as a nucleic acid?
> phenixbb mailing list
> phenixbb at phenix-online.org
More information about the phenixbb