Hi -
If you have a model 100% identical to your new complex, simply run Phaser, no need for Rosetta. I have a feeling that the second solution you get is simply a result from a slightly different model in the procedure.
I would expect a single solution from running phaser with your 3NAI as a model. I would then refine this structure as good as I could quickly with .refine, and inspect the maps. If protein B would not be there, then most likely it's not there in the crystal. If the crystal is similar to the protein without the protein b, I would had taken that as a sign that its unlikely to have the complex anyway.
A.
Sent from my iPad
On 16 Aug 2011, at 10:24, intekhab alam
Hi I solved an RNA polymerase structure in its native as well as in complex with chemical ligands (2.5A and C2 space group). The structure has been solved by MR a protein having 60% identity. The ASU has a monomer and dimer.
Recently i complexed this RNA polymerase (protein A) with a highly flexible protein B and got the crystal. These crytsal also have C2 space group (resolution 2.3A) having similar cell parameters but has psudotranslation (17.34%). I tried MR using rosetta. HHpred server selected one of my deposited structure in PDB (3NAI 100% identity) as template. Rosetta MR gave 2 solutions one of which is dimer-monomer arrangement while the other one is a trimer. The values of these solution are below.
Solution 1: trimer REMARK Log-Likelihood Gain: 10662.548 REMARK RFZ=16.5 TFZ=14.4 PAK=0 LLG=959 TFZ==23.2 RFZ=0.0 TFZ=48.9 PAK=0 LLG=3195 TFZ==55.6 (& TFZ==52.6) LLG+=(3191 & 7008) LLG=7024 TFZ==72.0 LLG=10663 TFZ==90.1 Solution 2: Monomer and dimer in ASU
REMARK Log-Likelihood Gain: 2120.965 REMARK RFZ=39.7 TFZ=21.9 PAK=0 LLG=811 TFZ==37.7 LLG=2121 TFZ==43.6 REMARK ENSEMBLE ensemble_1 EULER 204.07 178.91 204.37 FRAC 0.499
Map and model fits well in both the cases with R and R free 26 and 30% respectively.
My queries: 1. which one will be the correct solution? Is it possible that the protein B has resulted in the trimeric arrangement? 2. i cannot find any continuous 2Fo-Fc map for proteinB, but there are some continuous density at 0.7 which is very ambiguous. how can i improve the map so that i can model some fragments of my proteinB (protein B is of 73 amino acids and is predicted to be highly flexible and disordered. ((biochemical as well as bioinformatical analysis done ))
I already have tried autobuild but it fails to build even a partial fragment of the proteinB.
Kindly suggest how to proceed , or how can i prove the presence of protein B in the structure.
Regards INTEKHAB ALAM LABORATORY OF STRUCTURAL BIOINFORMATICS KOREA UNIVERSITY, SEOUL _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb