Re: [modeller_usage] modeling homodimer with hetero atoms
To: "" <>
Subject: Re: [modeller_usage] modeling homodimer with hetero atoms
From: NUTAN CHAUHAN <>
Date: Sat, 19 May 2012 00:28:16 +0800 (SGT)
Reply-to: NUTAN CHAUHAN <>
Many thanx for quick response. This problem is solved..actually problem was in script..i just needed to put "env.io.water=True" to allow water as modeller was not reading water as hetatm and reading it as alignment mismatch with pdb. Its done now. thanx a lot.
One more query related to this is:
Is there any way to assign the residue numbers starting from 1 to each of the chain in the model? actually when I run modeller to model dimer, it assigns numbers to the residues of chain A and continues to chainB like if chain a get finished at 100 then it starts chain B from 101.
thnx in advance.
regards,
Nutan
Chauhan Research Scholar
Department of Biotechnology BIT, Mesra Ranchi-835215, Jharkhand
From: NUTAN CHAUHAN <> To: "" <> Sent: Friday, 18 May 2012 1:50 PM Subject: modeling homodimer with hetero atoms
Dear all,
I am trying to model a homodimer with heteroatoms from a homodimer template. when make homodimer without heteroatoms it goes fine but when i try to build homodimer with heteroatoms it doesn't run and shows some errors regarding alignment mistakes.
My template protein consists of number of Waters, and other residues. All the htatms are given at the end of A and B chains and given the chain ids according to their chains they are attached with. I have removed all waters except
three (1 is atached to A and 2 with B
chain). So for now I have 7 and 8 hetatm on chain A and B respectively
What should be the alignment look like? I am giving 7 dots at the end of A chain and 8 dots at the end of B chain as followes:
Probably you overdid it with minimization. Try a smoother energy minimization scheme, e.g. 500-1000 steps. Also bare in mind that molecular mechanics force fields are not designed to produce crystallographic quality structures. I would rather use Rosetta (or PyRosetta) or http://www.yasara.org/minimizationserver.htm if I wanted to improve atom clashes, dihedral angles, secondary structure, etc.
> Respected Sir, > > I have tried to minimize my pdb structure which
is modeled by Modeller9v8. > i use the Discovery Studio plateform to minimize this pdb structure by > applying CHARMm force field and applying steepest descent and after that > conjugate gradient algorithm. > but the problem here is that the procheck result is not good as earlier. > such as in Ramachandran plot 87% amino acids are in favorable region before > minimization but after minimization only 55% amino acids are in favorable > range...so plz sir suggest me some solutions.... > > Thanking you, > > Your Sincerely > > -- > Vishal > Research Scholar, > University of Delhi, > North Campus > Delhi-110007 > M: 91+9650736653 > > > _______________________________________________ > modeller_usage mailing list > " target="_blank"
href="mailto:"> > https://salilab.org/mailman/listinfo/modeller_usage > >
On a quick glance, I am assuming you want to keep the sugars. You can specify which ones they are instead of using a generic "." E.g. Mannose is 2 Look in C:\Program Files (x86)\Modeller9.10\modlib\restyp_accelrys.lib for the full list.
Also there may some over-riding issue that excludes them as they are only in one template. Maybe you need to include a "." Before the dashes at the end other sequences e.g. SRW.----, this to me indicates a break of the chain and more residues "expected"
I have been trying to include some hetatms in my models but so far all that seems to appear apart from the main chain is a few characters at the end of the modelled sequence. I know that there must be something obvious that I am doing wrong but cannot for the life
of me work out what it is!
I have modified the PDBs 2Z7X, 2Z80, 2Z82 and 3A79 to include only the chain of interest
(using PyMOL) and only in 2Z7X have I kept the HETATMs of interest. I have copied in my script and alignment below.
Any help that can be offered would be so very much appreciated!
Best wishes,
Tory
from modeller import * from modeller.automodel import *
>P1;btau-tlr2 sequence:btau-tlr2:::::TLR2:bovine
AAL16722|AF368419:2.00:-1.00 SSLSCDPTGVCDGHSRSLNSIPSGLTAGVKSLDLSNNDITYVGNRDLQRCVNLKTLRLGA NEIHTVEEDSFFHLRNLEYLDLSYNRLSNLSSSWFRSLYVLKFLNLLGNLYKTLGETSLF SHLPNLRTLKVGNSNSFTEIHEKDFTGLTFLEELEISAQNLQIYVPKSLKSIQNISHLIL HLKQPILLVDILVDIVSSLDCFELRDTNLHTFHFSEASISEMSTSVKKLIFRNVQFTDES FVEVVKLFNYVSGILEVEFDDCTHDGIGDFRALSLDRIRHLGNVETLTIRKLHIPQFFLF HDLSSIYPLTGRVKRVTIENSKVFLVPCLLSQHLKSLEYLDLSENLMSEETLKNSACKDA WPFLQTLVLRQNRLKSLEKTGELLLTLENLNNLDISKNNFLSMPETCQWPGKMKQLNLSS TRIHSLTQCLPQTLEILDVSNNNLDSFSLILPQLKELYISRNKLKTLPDASFLPVLSVMR ISRNIINTFSKEQLDSFQQLKTLEAGGNNFICSCDFLSFTQGQ....* -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://salilab.org/archives/modeller_usage/attachments/20120430/530792b3/attachment.html>