Hi Eric,

 

Thank you so much for the response!

 

You are right! I didn’t use the alignment file. I directly uploaded a template structure in pdb and associated a target sequence with it.

 

Can you provide me some guidance on the correct way to do it? I’m guessing if I just associated the target sequence with the structure directly, Chimera would automatically generate an alignment for it before modeling the structure?

If so, is this the correct way to do it?

 

I saw a tutorial provided https://www.cgl.ucsf.edu/chimera/current/docs/UsersGuide/frametut.html.

In the Comparative Modeling Tutorial section, I think the alignment is generated after doing the blast search to find a template structure?

But in my case, I already have a pdb structure that I want to use as a template, so I am not doing a blast search to find the template. Is there a way I can obtain the alignment in Chimera in this case?

 

Thank you so much for your help!

 

Best,

Zoey

 

 

 

 

 

From: Eric Pettersen <pett@cgl.ucsf.edu>
Date: Tuesday, November 28, 2023 at 2:32 PM
To: Yang, Li-Yen <zoeyxyang@gatech.edu>
Cc: chimera-users@cgl.ucsf.edu <chimera-users@cgl.ucsf.edu>
Subject: Re: [Chimera-users] [chimera] modeller.SequenceMismatchError

Upon discussions with some of my colleagues, we're guessing you didn't even use an alignment file -- you just associated the structure with the target sequence directly and then ran Modeller.  That doesn't work if the target sequence has a deletion relative to the template sequence.  So until we fix that just use the alignment file I sent.

--Eric

> On Nov 28, 2023, at 10:49 AM, Eric Pettersen <pett@cgl.ucsf.edu> wrote:
>
> Hi Zoey,
>        The alignment you use has to have the actual sequence of the template structure.  In your alignment file, the four characters starting at position 81 of the template sequence are KMPK, whereas the corresponding part of the actual template structure is KMMSPK.  If I insert the missing "MS" into the template sequence and corresponding gap characters in the target sequence, then Modeller works.  I will send an edited alignment file with those changes in a separate email.
>
> --Eric
>
>        Eric Pettersen
>        UCSF Computer Graphics Lab
>
>