[phenixbb] Autobuild for very elongated models

Gino Cingolani Gino.Cingolani at jefferson.edu
Wed Oct 14 06:57:25 PDT 2009

Hi Everyone,

I am trying to solve a very elongated structure (~180A-long) formed by two long a-helices. I have a good MR solution (Log-Likelihood Gain ~168.537 to 3.0A resolution), which refines OK immediately after MR (Rfactor/Rfree 46/50%). The electron density has clear side chain features and high res data is available to ~2.1A resolution.

Logically, I should be able to plug everything into Autobuild and get a good final model (or at least something better than my input MR solution). I have tried a million time using different parameters and datasets, but Autobuild always breaks up my long helical model into short helical bundles, that refine to random Rfree. In other words, if my starting MR model is a continuous
80aa-long helix, Autobuild fragments it into three helices of ~30aa each packed laterally (I see back my initial model if I visualize symmetry-related atoms in Coot).

In parallel, I have made the following observation.
If I use DM to improve my phases, prior to Autobuild, I noticed that something similar happens. DM gives great density for less than half of my long helix while the rest is flattened. 
If I look at the solvent mask generated by DM, it basically flattens half of my elongated model. It seems that at least DM fails to generate a solvent mask that makes sense with such an elongated structure (by the way, the unit cell is a=25, b=37, c=220 in P212121). This problem can be circumvented in DM where 
an outside solvent mask can be inputted, but what about in phenix? 

Is it possible that Autobuild is failing because Resolve also makes a poor solvent mask? Is there a way to see and/or edit the solvent mask generated by Resolve? Or even better, can I input a new solvent mask in Autobuild?

Thanks in advance for the feedback!


Gino Cingolani, Ph.D.
Associate Professor
Thomas Jefferson University
Dept. of Biochemistry & Molecular Biology
233 South 10th Street - Room 826
Philadelphia PA 19107
Office (215) 503 4573
Lab    (215) 503 4595
Fax    (215) 923 2117
E-mail:   gino.cingolani at jefferson.edu
"Nati non foste per viver come bruti, ma per seguir virtute e conoscenza"
("You were not born to live like brutes, but to follow virtue and knowledge")
Dante, The Divine Comedy (Inferno,  XXVI, vv. 119-120) 

More information about the phenixbb mailing list