[Chimera-users] matching peptides with altenate locations

[Eric Pettersen]

Hi Sumitro,
	The problem is that there is no intrinsic ordering of atoms after an  
intersection operation.  Do you keep the order that was on the left  
side of the intersection or the order that was on the right side?   
Since the orders can differ, Chimera simply carries out the  
intersection operation in the fastest way possible with no regard to  
atom ordering.  Nonetheless, it seems natural to try to preserve the  
ordering on the left side of the intersection.  I will open a request  
in our Trac database for this feature, with you on the recipient list  
so you'll know when it gets implemented.
	In the interim, the only option you have is to be more explicit in  
the atom specifier.  This command gets the match you want:

match #0:1-6.c at ca:7.c at ca.a:8-9.c at ca #1:1.c at ca:2.c at ca.a:3.c at ca:4.c at ca.a: 
5-9.c at ca

With the two structures you mentioned, the RMSD from the above is  
0.087 angstroms.

--Eric

	Eric Pettersen
	UCSF Computer Graphics Lab

On Dec 31, 2009, at 2:12 AM, Sumitro Harjanto wrote:

> Hi elaine,
>
> I was experimenting with the "match" command and I found that when
> alternate locations are involved, sometimes the best alignment is not
> achieved.
>
> For example I had 3FQW as #0 and 3FQX as #1; I was comparing the  
> Chain C
> (9-mer peptide) of the two models.
> #0 had alternate location in the 7th residue while
> #1 had the alternate locations in the 2nd and 4th residue.
> They were essentially identical in sequence except for the 4th  
> residue.
>
> When I matched the 6 residues, which did not have any alternate
> locations, by the command:
>
> match #0:1.c,3.c,5.c,6.c,8.c,9.c at ca #1:1.c,3.c,5.c,6.c,8.c,9.c at ca
>
> the RMSD returned is 0.080, which was reasonable judging from the high
> conservation of the sequences.
>
> but when I matched all the residues, and designate the .b alternate
> location to be excluded by the command:
>
> match #0:.c at ca&~@.b #1:.c at ca&~@.b
>
> The RMSD returned was 8.730, which was unreasonably high. It was also
> clear by visual inspection that the alignment was way off the best. I
> would expect that the RMSD returned by the first and second command to
> be roughly equal; because the alternate locations of the residues do  
> not
> alter the C-alpha position by much, it was the side-chain positions  
> that
> were more extensively varied.
>
> Please advise. Thanks! =]
>
> Cheers,
> Sumitro
>
> -----Original Message-----
> From: Elaine Meng [mailto:meng at cgl.ucsf.edu]
> Sent: 16 December 2009 3:23 AM
> To: Sumitro Harjanto
> Cc: chimera-users at cgl.ucsf.edu
> Subject: Re: [Chimera-users] matching peptides with altenate locations
>
> Hi Sumitro,
> OK, I have 1akj as model 0 and 2gtz as model 1.  The second command
> you gave will work if you remove the spaces before and after "|":
>
> match #0:1-5.c at ca #1:1-3.c at ca|#1:4-5.c at ca.a
>
> The space is used to separate the two sets of atoms.  The first
> command you gave wouldn't work because only some of the residues in #1
> have alternate locations.
>
> If you want to use all nine residues, here is one possibility:
>
> match #0:1-9.c at ca #1:1-9.c at ca&~@.b
>
> meaning ...and not alternate location B.  If there were more alternate
> locations (C,D,...) you would also need to exclude those.  I hope this
> helps,
> Elaine
> -----
> Elaine C. Meng, Ph.D.                          meng at cgl.ucsf.edu
> UCSF Computer Graphics Lab and Babbitt Lab
> Department of Pharmaceutical Chemistry
> University of California, San Francisco
>                      http://www.cgl.ucsf.edu/home/meng/index.html
>
> On Dec 14, 2009, at 11:23 AM, Sumitro Harjanto wrote:
>
>> Hi,
>> I would like to match two peptides; one with alternate conformations
>> in residue 4 and 5. I have tried:
>>
>> match   #0:1-9.c at ca   #1:1-9.c at ca.a
>>
>> match   #0:1-5.c at ca   #1:1-3.c at ca   |   #1:4-5.c at ca.a
>>
>> but both don't work and give me the "unequal number of atom chosen"
>> error message.
>>
>> I tried changing match to sel in the latter case and I can get 10
>> atoms; 5 from each model, which is what I intended.
>> How should I deal with this?
>> Thanks,
>> Sumitro
>
>
> No virus found in this incoming message.
> Checked by AVG - www.avg.com
> Version: 8.5.427 / Virus Database: 270.14.108/2565 - Release Date:
> 12/14/09 19:40:00
> _______________________________________________
> Chimera-users mailing list
> Chimera-users at cgl.ucsf.edu
> http://www.cgl.ucsf.edu/mailman/listinfo/chimera-users