Molmap incorporation of B-factors
Hi all, For the purposes of model validation, I’d like to analyze local correlation of my model with the experimental map (and color it on either/both the model or map). I figured the easiest way to do this is to use molmap to generate a synthetic map, then use vop local correlation to generate a correlation map which I can use to color either the model or map (using scolor or “Values at Atom Positions”). This works fine, and generates informative looking pictures, I just wanted to clarify exactly what molmap is doing, to make sure I understand it. Am I correct in thinking molmap does not use a per-atom temperature factor (or alter the width of the gaussian depending on the atomic B-factor)? Because if this is the case I guess the way I am doing things will exaggerate how bad the model is in regions where the B-factor is high and the density weak (which is not the worst thing in the world - better than having an unrealistically rosy view of one’s model!). Would it be possible to incorporate atomic B-factors into molmap calculations for Chimera 2? Or does anyone know of an external program that will perform a similar calculation (I know I can generate RSCC plots in Phenix which must be doing this internally, but I rather like displaying it on the model and map for ease of interpretation). Cheers, Oliver.
Would be nice to include B-factors in the molmap calculation of a density map from an atomic model. Currently adds a Gaussian for each atom and all Gaussians have the same width (standard deviation) computed from the specified resolution and sigmaFactor as described in the documentation. https://www.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/molmap.html <https://www.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/molmap.html> I’ve made a Chimera feature request for this — more likely to go into Chimera 2 than Chimera 1. https://plato.cgl.ucsf.edu/trac/chimera/ticket/14258 Tom
On Oct 29, 2015, at 8:58 AM, Oliver Clarke <olibclarke@gmail.com> wrote:
Hi all,
For the purposes of model validation, I’d like to analyze local correlation of my model with the experimental map (and color it on either/both the model or map). I figured the easiest way to do this is to use molmap to generate a synthetic map, then use vop local correlation to generate a correlation map which I can use to color either the model or map (using scolor or “Values at Atom Positions”).
This works fine, and generates informative looking pictures, I just wanted to clarify exactly what molmap is doing, to make sure I understand it.
Am I correct in thinking molmap does not use a per-atom temperature factor (or alter the width of the gaussian depending on the atomic B-factor)?
Because if this is the case I guess the way I am doing things will exaggerate how bad the model is in regions where the B-factor is high and the density weak (which is not the worst thing in the world - better than having an unrealistically rosy view of one’s model!).
Would it be possible to incorporate atomic B-factors into molmap calculations for Chimera 2? Or does anyone know of an external program that will perform a similar calculation (I know I can generate RSCC plots in Phenix which must be doing this internally, but I rather like displaying it on the model and map for ease of interpretation).
Cheers, Oliver. _______________________________________________ Chimera-users mailing list Chimera-users@cgl.ucsf.edu http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users
participants (2)
-
Oliver Clarke -
Tom Goddard