Hi,
I have just been using MODELLER for a couple weeks now and have a question that is hopefully an easy one.
I am trying to model proteins of the cytochrome oxidase (COX) complex from a variety of organisms. I followed the "basic tutorial" online and got great results for the first subunit in the complex (COX1). But, when I tried the same thing for COX2, the align2d.py script aligns my sequence against ChainA in the complex instead of ChainB, so I end up with something that looks like COX1 instead of COX2. This happens even though I change the "aln.append_model" line in align2d.py to align_codes='1v54B', instead of 1v54A.
I was able to work around this by following the advanced tutorial and using the multiple templates scripts (even though there is sometimes only a single template to compare the sequence to). However, I am not able to compare the correct DOPE profiles when trying plot_profiles.py; ChainA's profile is being used when comparing to the model instead of ChainB. I would like to be able to get an idea of how good the model is and identify areas of weak reliability, so the DOPE profiles seem like a good tool.
Any help would be appreciated, as I really like the MODELLER software so far and hope to be able to make use of it.
Thanks,
Justin
************************************************************ Justin C. Havird, PhD Department of Biology Colorado State University Room E106 Anatomy/Zoology Building Fort Collins, CO 80523 Lab telephone: 970-491-2058 Fax: 970-491-0649 Email: Justin.Havird@colostate.edu Justin's Research Website: http://jchavird.wix.com/jchavird Sloan Lab Website: https://sites.google.com/site/danielbsloan/ ************************************************************
On 10/22/2014 02:05 PM, Justin Havird wrote: > I am trying to model proteins of the cytochrome oxidase (COX) complex > from a variety of organisms. I followed the "basic tutorial" online and > got great results for the first subunit in the complex (COX1). But, when > I tried the same thing for COX2, the align2d.py script aligns my > sequence against ChainA in the complex instead of ChainB, so I end up > with something that looks like COX1 instead of COX2. This happens even > though I change the "aln.append_model" line in align2d.py to > align_codes='1v54B', instead of 1v54A.
It's hard to tell exactly what your problem is without seeing your script and alignment, but it sounds like you're not specifying model_segment when you read in your complex, so Modeller's reading in both chains, not just one.
Ben Webb, Modeller Caretaker