normalizing attributes through point mutant models
Hello I´m working with one antigen/antibody complexes and different point mutations at the interface. I can calculate and draw at the surface attributes like atomic clashes "overlap", but what I don´t know, is how to normalize the calculation output for 2 complexes, neither render such "normalized" attributes in the models. I have tried at Render attribute tool, the option "create corresponding color key", introducing specific clashes boundaries and their coloring values but there is no "apply" button, such the changes doesn't goes to the model. Any help or tip? Thanks in advance Ruben
Hello Ruben, With Find Clashes/Contacts, clashes can be shown with lines and/or by coloring the clashing atoms a single color: <http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/findclash/findclash.html> It also allows assigning the "overlap" attribute of atoms so that you can color individual atoms differently depending on the amount of overlap, using Render by Attribute. <http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/render/render.html#render> It sounds like you didn't understand how to change the values-to-colors mapping in Render by Attribute. This is *not* done in the Color Key tool. Instead it is done in the Render by Attribute tool: you can move the colored bars on the histogram directly with the mouse or by typing in values, change the colors of the bars, etc. as explained here: <http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/render/render.html#colors> After you have the desired value-to-color mapping already figured out in Render by Attribute, then you might draw a color key to include in a figure. For example, in this image, the color key is the rectangle below the structure: <http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/tutorials/2gbp-bfactor.png> Secondly, I don't understand what normalization you are trying to do, or why. It seems (to me) that the actual overlap values and counts of clashes would be more meaningful for comparison of the complexes. Best wishes, Elaine ----- Elaine C. Meng, Ph.D. UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco On May 28, 2014, at 11:15 AM, Rubén Marrero Diaz De Villegas wrote:
Hello I´m working with one antigen/antibody complexes and different point mutations at the interface. I can calculate and draw at the surface attributes like atomic clashes “overlap”, but what I don´t know, is how to normalize the calculation output for 2 complexes, neither render such “normalized” attributes in the models.
I have tried at Render attribute tool, the option “create corresponding color key”, introducing specific clashes boundaries and their coloring values but there is no “apply” button, such the changes doesn’t goes to the model. Any help or tip? Thanks in advance Ruben
Elaine Thanks for your opinion Now I understand, i was wrong about the usage of " create corresponding color key " Anyway, and as example, suppose I load 5 interaction complexes in chimera, the 5 complexes have the same AA sequence , differing only in one AA, then I need to weight, or evaluate in some way, the relevance of the point mutations in the interacting interface at each complex, and visualize and plot this calculation, Do I made such task in chimera ?? and HOW? remember, there is 5 interfaces with individual values that must be compared. For example, comparing the atomic clashes between interface residues ? Thanks in advance Ruben -----Mensaje original----- De: Elaine Meng [mailto:meng@cgl.ucsf.edu] Enviado el: Friday, May 30, 2014 7:16 PM Para: Rubén Marrero Diaz De Villegas CC: chimera-users@cgl.ucsf.edu Asunto: Re: [Chimera-users] normalizing attributes through point mutant models Hello Ruben, With Find Clashes/Contacts, clashes can be shown with lines and/or by coloring the clashing atoms a single color: <http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/findclash/findclash.html> It also allows assigning the "overlap" attribute of atoms so that you can color individual atoms differently depending on the amount of overlap, using Render by Attribute. <http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/render/render.html#render> It sounds like you didn't understand how to change the values-to-colors mapping in Render by Attribute. This is *not* done in the Color Key tool. Instead it is done in the Render by Attribute tool: you can move the colored bars on the histogram directly with the mouse or by typing in values, change the colors of the bars, etc. as explained here: <http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/render/render.html#colors> After you have the desired value-to-color mapping already figured out in Render by Attribute, then you might draw a color key to include in a figure. For example, in this image, the color key is the rectangle below the structure: <http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/tutorials/2gbp-bfactor.png> Secondly, I don't understand what normalization you are trying to do, or why. It seems (to me) that the actual overlap values and counts of clashes would be more meaningful for comparison of the complexes. Best wishes, Elaine ----- Elaine C. Meng, Ph.D. UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco On May 28, 2014, at 11:15 AM, Rubén Marrero Diaz De Villegas wrote:
Hello I´m working with one antigen/antibody complexes and different point mutations at the interface. I can calculate and draw at the surface attributes like atomic clashes "overlap", but what I don´t know, is how to normalize the calculation output for 2 complexes, neither render such "normalized" attributes in the models.
I have tried at Render attribute tool, the option "create corresponding color key", introducing specific clashes boundaries and their coloring values but there is no "apply" button, such the changes doesn't goes to the model. Any help or tip? Thanks in advance Ruben
Hi Ruben, I don't understand your wish to "normalize". If I were comparing the different structures, I would just get the count of how many clashes (or H-bonds or whatever you are looking at) there are in each structure, perhaps for the whole interface and/or just involving the one residue that is different, and simply report those numbers. Maybe you want to divide by something, like number of residues or surface area of interface. However, you didn't say what the "something" is, so I have no way of answering whether you can measure that "something" or not. Even if you can measure it (for example, "measure buriedArea" command <http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/measure.html> ) if I were trying to do something like that I would just do the 5 measurements and the 5 divisions "manually" (like with a calculator). Best, Elaine ---------- Elaine C. Meng, Ph.D. UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco On Jun 5, 2014, at 8:16 AM, Rubén Marrero Diaz De Villegas <marrerodiaz.ruben@inta.gob.ar> wrote:
Elaine
Thanks for your opinion Now I understand, i was wrong about the usage of " create corresponding color key " Anyway, and as example, suppose I load 5 interaction complexes in chimera, the 5 complexes have the same AA sequence , differing only in one AA, then I need to weight, or evaluate in some way, the relevance of the point mutations in the interacting interface at each complex, and visualize and plot this calculation, Do I made such task in chimera ?? and HOW? remember, there is 5 interfaces with individual values that must be compared. For example, comparing the atomic clashes between interface residues ?
Thanks in advance
Ruben
participants (2)
-
Elaine Meng -
Rubén Marrero Diaz De Villegas