[3dem] Re: 3dem Digest, Vol 10, Issue 5

Weimin Wu wu49 at purdue.edu
Fri Jun 13 14:26:01 PDT 2008


3dem-request at ncmir.ucsd.edu wrote:
> Send 3dem mailing list submissions to
> 	3dem at ncmir.ucsd.edu
>
> To subscribe or unsubscribe via the World Wide Web, visit
> 	https://mail.ncmir.ucsd.edu/mailman/listinfo/3dem
> or, via email, send a message with subject or body 'help' to
> 	3dem-request at ncmir.ucsd.edu
>
> You can reach the person managing the list at
> 	3dem-owner at ncmir.ucsd.edu
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of 3dem digest..."
>
>
> Today's Topics:
>
>    1. Do you have the program "findort" in IMIRS. (wanggongming)
>    2. Particle number in a class to produce an average (daixh)
>    3. Re: Re: [3dem] Particle number in a class to produce an
>       average (daixh)
>    4. CEMOVIS course announcement (Sergej Masich)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Fri, 13 Jun 2008 09:53:40 +0800
> From: "wanggongming" <wanggongming at ict.ac.cn>
> Subject: [3dem] Do you have the program "findort" in IMIRS.
> To: "3dem at ncmir.ucsd.edu" <3dem at ncmir.ucsd.edu>
> Message-ID: <200806130953401407253 at ict.ac.cn>
> Content-Type: text/plain; charset="gb2312"
>
> Dear 3dem friends: 
>     I use IMIRS to reconstruct 3D-structure of proteins recently. As everybody knows that "findort" program is very important in IMIRS software package. We can select initial templates from the original center and orientation files created by "ortall" program. The result of templates are recorded in the "record.txt" file. 
>     But I could not find the "findort" program. I don't know why it is not in the IMIRS softwre package. It is likely that this program has renamed in this version that I downloaded. Do you have the "findort" program or have the experience in the area of selecting initial templates. Could you send this program to me if you have or give me some suggestion how to solve the seclecting problem.  
>     Sincerely
>     GongMing Wang
>
>
>
> wanggongming
> 2008-06-13
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: https://mail.ncmir.ucsd.edu/mailman/private/3dem/attachments/20080613/95d6a76b/attachment-0001.html
>
> ------------------------------
>
> Message: 2
> Date: Fri, 13 Jun 2008 10:49:49 +0800
> From: "daixh" <daixh at mail2.sysu.edu.cn>
> Subject: [3dem] Particle number in a class to produce an average
> To: "3dem at ncmir.ucsd.edu" <3dem at ncmir.ucsd.edu>
> Message-ID: <200806131049497031500 at mail2.sysu.edu.cn>
> Content-Type: text/plain; charset="gb2312"
>
> Hi everyone!
> Anyone familiar with the single particle reconstruction knows how many particles in a class are sufficient to produce averages for a reconstruction about 8 Angstroms? I know it depends on the data quality, but what is the typical number or range? 
> Thanks.
>
>
>
>
> Dai Xinghong
> Laboratory of BioEM
> School of Life Science
> Sun Yat-sen University
> Guangzhou, China. 510275
> daixh at mail2.sysu.edu.cn
> http://bioem.sysu.edu.cn
> Tel:+86 20 84112286
> Fax:+86 20 84110108
>
> 2008-06-13
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: https://mail.ncmir.ucsd.edu/mailman/private/3dem/attachments/20080613/47fd16cd/attachment-0001.html
>
> ------------------------------
>
> Message: 3
> Date: Fri, 13 Jun 2008 12:54:01 +0800
> From: "daixh" <daixh at mail2.sysu.edu.cn>
> Subject: Re: Re: [3dem] Particle number in a class to produce an
> 	average
> To: "3dem" <3dem at ncmir.ucsd.edu>
> Message-ID: <200806131254015462793 at mail2.sysu.edu.cn>
> Content-Type: text/plain; charset="gb2312"
>
> ·¢¼þÈË£º daixh
> ·¢ËÍʱ¼ä£º 2008-06-13 12:52:18
> ÊÕ¼þÈË£º Steven Ludtke
> Ö÷Ì⣺ Re: Re: [3dem] Particle number in a class to produce an average
>
> Yeah, I am doing the reconstruction with EMAN1.8. The particles are about 460pixs with 1.06A/pix, boxed out with 512*512 boxes. The data are acquired on the JEM3100F, and the quality is pretty good. 
> 18054 particles applied with D5 symmetry are reconstructed, and the result is measured to be ~8A by the FSC=0.5 criteria. However, the structure itself seems to be far away from 8A resolution.(If it is turely 8A, we should more or less be able to see some 2nd structure elements, right?) I checked the particle numbers in each class. It seems that partial orientation preference exists and classes near the "side wiew" and "top view" have particles up to ~100, while many intermediate classes have <10 particles(~5-10) (However, these class averages are not eliminated by most stringent parameter settings in "refine" program of EMAN).  So, is this most probablely the reason? 
> And to what extend can we trust the FSC curve? ...
> Thanks for your patience.
>
>
>
>
> Dai Xinghong
> Laboratory of BioEM
> School of Life Science
> Sun Yat-sen University
> Guangzhou, China. 510275
> daixh at mail2.sysu.edu.cn
> http://bioem.sysu.edu.cn
> Tel:+86 20 84112286
> Fax:+86 20 84110108
>
> 2008-06-13
>
>
>
> ·¢¼þÈË£º Steven Ludtke
> ·¢ËÍʱ¼ä£º 2008-06-13 10:52:20
> ÊÕ¼þÈË£º daixh
> ³­ËÍ£º 
> Ö÷Ì⣺ Re: [3dem] Particle number in a class to produce an average
>
> There is no 'typical number' it depends VERY strongly on particle size, box size, MTF, 
> CTF, and, frankly, the software.  A general rule of thumb for cryo data might be ~10-20
> particles per class at a minimum if you are doing an EMAN or IMAGIC style reconstruction.
> However, the number of classes will vary...
>
> ----------------------------------------------------------------------------
> Steven Ludtke, PhD              |        Baylor College of Medicine
> sludtke at bcm.tmc.edu             |     Associate Professor & Co-Director
> stevel at alumni.caltech.edu       | National Center For Macromolecular Imaging 
> V: (713)798-9020                |    Dept of Biochemistry and Mol. Biol.
> F: (713)798-1625                |
>                                 |             Those who Do, Are      
> http://ncmi.bcm.edu/~stevel     |         The converse also applies   
>
>
>
>
> _______________________________________________
> 3dem mailing list
> 3dem at ncmir.ucsd.edu
> https://mail.ncmir.ucsd.edu/mailman/listinfo/3dem
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: https://mail.ncmir.ucsd.edu/mailman/private/3dem/attachments/20080613/c4ca07ee/attachment-0001.html
>
> ------------------------------
>
> Message: 4
> Date: Fri, 13 Jun 2008 18:37:57 +0200
> From: Sergej Masich <Sergej.Masich at ki.se>
> Subject: [3dem] CEMOVIS course announcement
> To: 3dem at ncmir.ucsd.edu
> Message-ID: <71AA7D33-117D-4752-9209-118DD11ABAEC at ki.se>
> Content-Type: text/plain; charset=WINDOWS-1252; format=flowed;
> 	delsp=yes
>
> Dear all,
>
> Department of Cell and Molecular Biology at Karolinska Institutet,  
> Stockholm is pleased to announce a course on “CEMOVIS: cryo-electron  
> microscopy of vitreous sections”. More information as well as an  
> application form can be found at http://www.3dem-noe-training.org/
>
> Best regards,
>
> Sergej Masich, Ph.D.
> ==============================
> Dr. Sergej Masich
> Dept. of Cell and Molecular Biology
> Karolinska Institutet
> Box 285
> 171 77 Stockholm, Sweden
> tel: (+)46 - 8 - 524 873 61
> mobile: (+)46 - 736 - 833 693
> fax: (+)46 - 8 - 31 35 29
> e-mail: Sergej.Masich at ki.se
>
>
>
>
>
>
> ------------------------------
>
> _______________________________________________
> 3dem mailing list
> 3dem at ncmir.ucsd.edu
> https://mail.ncmir.ucsd.edu/mailman/listinfo/3dem
>
>
> End of 3dem Digest, Vol 10, Issue 5
> ***********************************
>   
There are a lot of tricky things to get a correct resolution evaluation. 
As you said the structure itself is far away from 8A as determined in 
FSC curve. I would like to say in your alignment you might use too tight 
mask or in your postprocess of your 3D maps, you might use very 
aggressive mask. And it is true when you choose box size 512 for the 
460-pixel particle. You might choose 600 or 640 as box size. The uneven 
distributed orientation is true in phage sample, for D5 symmetry, I 
don't know what is the angle step size you choose, but for getting 8A, 
1.2degree might be enough. The average number in each class will be 
around 35. Therefore in some class there might be less than 10 
particles. You might choose make3d from individual particles which has 
similar quality to the classesavg result.

weimin
Purdue Univ.


More information about the 3dem mailing list