Hi Zhihai,

  My mistake, the subdivision of an icosahedron I suggested to you is a T=9 lattice not the T=7 lattice you want.  The technique I suggested cannot produce a T=7 lattice.  What you want is the h=1,k=2 lattice made by “shape icos radius 100 lattice 1,2”, but instead of showing pentagons and hexagons you want the complementary lattice where the center of each pentagon and hexagon is a vertex and those vertices are connected to make triangles.  As I mentioned before, Chimera can’t make that complementary lattice.  I looked at the code and don’t see a simple way to produce the lattice you want. But I also think that the figure you sent would probably be clearer using the pentagon/hexagon lattice as Chimera displays them.  In the figure from the PNAS paper with triangle lattice it is hard to see then pentagonal vertices in the lattice (they are colored gray instead of yellow) and the lattice lines overlap and obscure the squigly lines indicating the interactions between capsomeres.  So I feel it would be clearer to show the pentagon and hexagon cage.  You may disagree.  While I could code the lattice of triangles that you want, I think it is rarely used (no-one requested it in the last 5 years) and it would take me considerable time to implement, probably a full day.  Because I have many other more commonly used features to work on I don’t think I can do this one now.

Tom



From: 李智海 
Subject: Re: [Chimera-users] Icosahedral model
Date: June 19, 2014 at 8:42:27 PM PDT
To: Tom Goddard 
Great!  I got that model, thank you!
But I have a further question. It seems to that this model doesn’t represent the capsid of EMDB5155. it have more than 72 vertex including the vertex on the middle of every icosahedral face, isn’t it? Moreover, the arrangement here is different from the capsid because no pentamers of the capsid is located at icosahedral edges.
Do you have any ideas?
Thanks!

Zhihai Li

On Jun 19, 2014, at 11:19 AM, Tom Goddard  wrote:

Hi Zhihai,

  Since you made the icosahedron radius 300 compared to the example value I used of 100, you also need to increase the “spacing” parameter used to subdivide the icosahedron from 50 to 150 (3 times larger).  The spacing value controls how many segments each edge is subdivide into — it is the maximum length of a subdivided edge.

Tom


Begin forwarded message:

From: 李智海
Subject: Re: [Chimera-users] Icosahedral model
Date: June 19, 2014 at 1:14:36 AM PDT
To: Tom Goddard 

<image1.png>
<image2.png>Hi, Tom,
Thanks for your quick reply!
Here I have a problem that I can’t repeat your work. 

After entering the commands:
shape icos radius 300 orientation 222r linewidth 3 color red mesh true div 1
sop finerMesh #0 spacing 50 

I got a different model as a result (see attached figures). 
Do you have any ideas? Thanks !

Zhihai


在 2014年6月19日，上午2:38，Tom Goddard  写道：

Hi Zhihai,

  The example icosahedral cage figure you sent is unusual because it has virus pentamers at six-fold vertices.  The arrangement is a T = 7d lattice and the Chimera command “shape icos radius 300 lattice 1,2” makes a matching cage with hexagons and pentagons as shown in the attached image.  But I can see how you don’t want that because the cage hexagons have pentamers in the middle.  To make a cage like the figure you show is tricky in Chimera.  It is basically an icosahedron (12 vertices and 20 faces) with each edge subdivided in 3 parts.  Here’s a way to make such an subdivision in Chimera

shape icos radius 100 mesh true div 1
sop finerMesh #0 spacing 50

The first command makes a 12 vertex icoshedron, and the second command subdivides it to make a 72 vertex cage.  The result has pointy 5-fold vertices.  To make it more spherical as in the figure you sent you can smooth it with a command I just added today

sop smooth #1 iter 8 inplace true

This “sop smooth” command moves the vertices towards their connected neighbors and will be in tonight’s Chimera daily build.  I’ve attached an image of the result after using the meshmol command to replace the edge lines with cylinders.

Tom

<emd5155_cage.jpg><icos72.jpg>

From: 李智海 
Subject: Re: [Chimera-users] Icosahedral model
Date: June 17, 2014 at 7:38:03 PM PDT
To: Tom Goddard 


Hi, Tom,<无标题.png>
Thanks very much for your reply! 
However, it seems to be that I can’t achieve what I want by performing these processes. Every capsomere should be a node in the icosahedral model which represents the distribution of capsomeres on the virus capsid.
Attached is an example model (right) of what I want from an article(Mattias Wolf et al. PNAS ) .  
Can I get that model from Chimera?
Any help would be appreciated.

Best wishes!

Zhihai Li


在 2014年6月11日，下午11:52，Tom Goddard <goddard@sonic.net> 写道：

Hi Zhihai,

  You can draw an icosahedron mesh with the Icosahedron Surface tool (menu Tools / Higher-Order Structure), or the Chimera “shape” command, for example,

shape icosahedron radius 55 orientation 2n5 lattice 2,1

You’ll need to figure out the right radius (just try different values and compare the size to the density map), the density map should be centered (use Volume Viewer menu Features / Coordinates to set the origin for the map), figure out which of the standard icosahedron orientations your map uses (just try different ones and compare to your map), and figure out the two lattice parameters if you want to replicate the protein packing in the capsid.  The graphical Icosahedron Surface tool does not let you set lattice parameters — it only shows a simple 12 vertex icosahedron.

  If you want to turn the mesh into a ball and stick model you use the Chimera “meshmol” command.

  Here is documentation of these commands





  Tom

<em_1f8v.png>


On Jun 10, 2014, at 8:48 PM, 李智海 <21620101152414@stu.xmu.edu.cn> wrote:

Hi，
I have a density map of an icosahedral capsid, and now i want to make a icosahedral model to represent the capsid. In this model, every corner represent a capsomere. what can i do to display this in Chimera ? Do I require any plug-in?
Thank you！
Best wishes!

Zhihai Li

School of Life Sciences, Xiamen University.
Xiang'an Road(South), Xiang'an District, Xiamen City,
Fujian Province, China.

_______________________________________________
Chimera-users mailing list
Chimera-users@cgl.ucsf.edu
http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users


On Jun 11, 2014, at 8:52 AM, Tom Goddard <goddard@sonic.net> wrote:

Hi Zhihai,

  You can draw an icosahedron mesh with the Icosahedron Surface tool (menu Tools / Higher-Order Structure), or the Chimera “shape” command, for example,

shape icosahedron radius 55 orientation 2n5 lattice 2,1

You’ll need to figure out the right radius (just try different values and compare the size to the density map), the density map should be centered (use Volume Viewer menu Features / Coordinates to set the origin for the map), figure out which of the standard icosahedron orientations your map uses (just try different ones and compare to your map), and figure out the two lattice parameters if you want to replicate the protein packing in the capsid.  The graphical Icosahedron Surface tool does not let you set lattice parameters — it only shows a simple 12 vertex icosahedron.

  If you want to turn the mesh into a ball and stick model you use the Chimera “meshmol” command.

  Here is documentation of these commands





  Tom

<em_1f8v.png>


On Jun 10, 2014, at 8:48 PM, 李智海 <21620101152414@stu.xmu.edu.cn> wrote:

Hi，
I have a density map of an icosahedral capsid, and now i want to make a icosahedral model to represent the capsid. In this model, every corner represent a capsomere. what can i do to display this in Chimera ? Do I require any plug-in?
Thank you！
Best wishes!

Zhihai Li

School of Life Sciences, Xiamen University.
Xiang'an Road(South), Xiang'an District, Xiamen City,
Fujian Province, China.

_______________________________________________
Chimera-users mailing list
Chimera-users@cgl.ucsf.edu
http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users



On Jun 18, 2014, at 11:38 AM, Tom Goddard <goddard@sonic.net> wrote:

Hi Zhihai,

  The example icosahedral cage figure you sent is unusual because it has virus pentamers at six-fold vertices.  The arrangement is a T = 7d lattice and the Chimera command “shape icos radius 300 lattice 1,2” makes a matching cage with hexagons and pentagons as shown in the attached image.  But I can see how you don’t want that because the cage hexagons have pentamers in the middle.  To make a cage like the figure you show is tricky in Chimera.  It is basically an icosahedron (12 vertices and 20 faces) with each edge subdivided in 3 parts.  Here’s a way to make such an subdivision in Chimera

shape icos radius 100 mesh true div 1
sop finerMesh #0 spacing 50

The first command makes a 12 vertex icoshedron, and the second command subdivides it to make a 72 vertex cage.  The result has pointy 5-fold vertices.  To make it more spherical as in the figure you sent you can smooth it with a command I just added today

sop smooth #1 iter 8 inplace true

This “sop smooth” command moves the vertices towards their connected neighbors and will be in tonight’s Chimera daily build.  I’ve attached an image of the result after using the meshmol command to replace the edge lines with cylinders.

Tom

<emd5155_cage.jpg><icos72.jpg>

From: 李智海 
Subject: Re: [Chimera-users] Icosahedral model
Date: June 17, 2014 at 7:38:03 PM PDT
To: Tom Goddard 


Hi, Tom,<无标题.png>
Thanks very much for your reply! 
However, it seems to be that I can’t achieve what I want by performing these processes. Every capsomere should be a node in the icosahedral model which represents the distribution of capsomeres on the virus capsid.
Attached is an example model (right) of what I want from an article(Mattias Wolf et al. PNAS ) .  
Can I get that model from Chimera?
Any help would be appreciated.

Best wishes!

Zhihai Li


在 2014年6月11日，下午11:52，Tom Goddard <goddard@sonic.net> 写道：

Hi Zhihai,

  You can draw an icosahedron mesh with the Icosahedron Surface tool (menu Tools / Higher-Order Structure), or the Chimera “shape” command, for example,

shape icosahedron radius 55 orientation 2n5 lattice 2,1

You’ll need to figure out the right radius (just try different values and compare the size to the density map), the density map should be centered (use Volume Viewer menu Features / Coordinates to set the origin for the map), figure out which of the standard icosahedron orientations your map uses (just try different ones and compare to your map), and figure out the two lattice parameters if you want to replicate the protein packing in the capsid.  The graphical Icosahedron Surface tool does not let you set lattice parameters — it only shows a simple 12 vertex icosahedron.

  If you want to turn the mesh into a ball and stick model you use the Chimera “meshmol” command.

  Here is documentation of these commands





  Tom

<em_1f8v.png>


On Jun 10, 2014, at 8:48 PM, 李智海 <21620101152414@stu.xmu.edu.cn> wrote:

Hi，
I have a density map of an icosahedral capsid, and now i want to make a icosahedral model to represent the capsid. In this model, every corner represent a capsomere. what can i do to display this in Chimera ? Do I require any plug-in?
Thank you！
Best wishes!

Zhihai Li

School of Life Sciences, Xiamen University.
Xiang'an Road(South), Xiang'an District, Xiamen City,
Fujian Province, China.

_______________________________________________
Chimera-users mailing list
Chimera-users@cgl.ucsf.edu
http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users


On Jun 11, 2014, at 8:52 AM, Tom Goddard <goddard@sonic.net> wrote:

Hi Zhihai,

  You can draw an icosahedron mesh with the Icosahedron Surface tool (menu Tools / Higher-Order Structure), or the Chimera “shape” command, for example,

shape icosahedron radius 55 orientation 2n5 lattice 2,1

You’ll need to figure out the right radius (just try different values and compare the size to the density map), the density map should be centered (use Volume Viewer menu Features / Coordinates to set the origin for the map), figure out which of the standard icosahedron orientations your map uses (just try different ones and compare to your map), and figure out the two lattice parameters if you want to replicate the protein packing in the capsid.  The graphical Icosahedron Surface tool does not let you set lattice parameters — it only shows a simple 12 vertex icosahedron.

  If you want to turn the mesh into a ball and stick model you use the Chimera “meshmol” command.

  Here is documentation of these commands





  Tom

<em_1f8v.png>


On Jun 10, 2014, at 8:48 PM, 李智海 <21620101152414@stu.xmu.edu.cn> wrote:

Hi，
I have a density map of an icosahedral capsid, and now i want to make a icosahedral model to represent the capsid. In this model, every corner represent a capsomere. what can i do to display this in Chimera ? Do I require any plug-in?
Thank you！
Best wishes!

Zhihai Li

School of Life Sciences, Xiamen University.
Xiang'an Road(South), Xiang'an District, Xiamen City,
Fujian Province, China.

_______________________________________________
Chimera-users mailing list
Chimera-users@cgl.ucsf.edu
http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users