Dear Almudena,
I hope you have indeed solved the structure, but any model that has been fit and refined to the diffraction data will give a very strong signal in Phaser. I’m afraid that running molecular replacement in Phaser is not an independent check for the validity of the model — unless of course you have a second crystal form, in which case overfitting of the data in the first crystal form wouldn’t affect the fit to the data for the second form.
Best wishes,
Randy
-----
Randy J. Read
Department of Haematology, University of Cambridge
Cambridge Institute for Medical Research Tel: +44 1223 336500
Wellcome Trust/MRC Building Fax: +44 1223 336827
Hills Road E-mail: [log in to unmask]
Cambridge CB2 0XY, U.K. www-structmed.cimr.cam.ac.uk
> On 30 Jul 2015, at 11:36, Almudena Ponce Salvatierra <[log in to unmask]> wrote:
>
> Hi David, Hi Bert,
>
> thanks for getting back so quickly. So, the longest stretches of autotraced protein that I have are around 100 residues. I am now running a refinement and also autobuild. In any case I will try resolve too, I have 4 copies of my protein in the ASU so the NCS averaging should work well there.
>
> Thanks a lot!! hopefully I will have some happy news soon!
>
> Best wishes,
>
> Almudena
>
> 2015-07-30 12:10 GMT+02:00 Bert Van-Den-Berg <[log in to unmask]>:
> If you have NCS, give the model to Autobuild within phenix. You can try either model building with resolve or resolve/buccaneer. The building itself won't be great probably (like you have now) due to the low resolution but the density modified maps produced by Resolve can be great even at lo-res (but you'd need to have NCS so that averaging produces better maps). This has worked great for me in the past.
>
> Maps obtained from refinement at this stage probably won't be useful; my experience is that Rfree needs to be below 40% or so. You probably will need to build by hand first to get Rfree to those kind of values.
>
> good luck, bert
> From: CCP4 bulletin board [[log in to unmask]] on behalf of Almudena Ponce Salvatierra [[log in to unmask]]
> Sent: Thursday, July 30, 2015 10:31 AM
> To: [log in to unmask]
> Subject: [ccp4bb] puzzled
>
> Dear all,
>
> I would like to ask your opinion on something (I was actually not expecting to work...but it did!!).
>
> I am now solving a protein of 135 kDa. I did SIRAS within SHELX and the poly ala trace I discarded, but I kept the density modified map.
>
> I gave the map and the sequence to buccaneer and it gave back 1532 residues built in 133 fragments. This looked more like a spaghetti dish (excuse me, I'm used not NA rather than proteins), with lots of small fragments... but I wanted to give it a try and gave it to Phaser for molecular replacement (here is where I was having the most doubts). I gave it along with a native data set collected at 2.9 Angstroms and phaser gives a TFZ score of 90.6 plus an LLG of 4512.88. So, it looks like it is a correct solution!!
>
> Now the question is, this is mostly a poly-ala thing, with lots of fragments... can I think that actually this is a solution? I want to, actually, I mean, at the end these short pieces are oriented respect to each other in a specific manner, even if they're not connected, and phaser could place it. But now my question is, how or from where do I start to make it continuous? Or to complete the structure...
>
> Any comments on this are more than welcome!!
>
> Best wishes,
>
> Almu
>
> --
> Almudena Ponce-Salvatierra
> Macromolecular crystallography and Nucleic acid chemistry
> Max Planck Institute for Biophysical Chemistry
> Am Fassberg 11 37077 Göttingen
> Germany
>
>
>
>
> --
> Almudena Ponce-Salvatierra
> Macromolecular crystallography and Nucleic acid chemistry
> Max Planck Institute for Biophysical Chemistry
> Am Fassberg 11 37077 Göttingen
> Germany
>
|