Dear All,
I'm currently trying to model a protein that has a thioester linkage between two residues. I've searched archives and Google and there's no mention on how to include such a linkage in the model. So the question is: is it possible? If so, how? If not, how can I overcome this?
Best regards to all and thanks for the answers,
João [ .. ] Rodrigues
(Blog) http://doeidoei.wordpress.com (MSN) always_asleep_@hotmail.com (Skype) rodrigues.jglm
João Rodrigues wrote: > I'm currently trying to model a protein that has a thioester linkage > between two residues. I've searched archives and Google and there's no > mention on how to include such a linkage in the model. So the question > is: is it possible? If so, how? If not, how can I overcome this?
As Joel points out, the way to do this is to create a custom patch residue. See http://salilab.org/modeller/FAQ.html#9 and http://www.charmm.org for more details on CHARMM patch residues. The link you posted describes how to create a new residue type, not how to make a link between two existing residues.
The example in the FAQ is for a disulfide, but I imagine your linkage would be similar. As you already discovered, this patch residue should go in modlib/top_heav.lib.
Ben Webb, Modeller Caretaker
Ok, I got it. I'm checking the pages and I think I'm gonna make it. Kind of helps to have the hopes high :)
Another question though, maybe this is a bit out of the scope of MODELLER but:
PRES DISU -0.36 ! Patch for disulfides. Patch must be 1-CYS and 2-CYS.
ATOM 1:CB CT2 -0.10 ! ATOM 1:SG SM -0.08 ! 2:SG--2:CB-- ATOM 2:SG SM -0.08 ! / ATOM 2:CB CT2 -0.10 ! -1:CB--1:SG
So, I get the syntax but I fail to understand one thing. Is that number -0.36 the overall charge of the residue or something? Because in the CHARMM Struct.doc file it says:
[ATOM <I><atomname> <parameter type> <charge> ]
But it doesn't mention anything next to the PRES line. Is this a MODELLER specific thing? What happens if I leave it blank? Is there a way to generate this automatically? Or are these set values written down somewhere?
Sorry for all the questions, but I'm trying to figure out how this works and not just copy paste things..
João [ .. ] Rodrigues
(Blog) http://doeidoei.wordpress.com (MSN) always_asleep_@hotmail.com (Skype) rodrigues.jglm
João Rodrigues wrote: > Another question though, maybe this is a bit out of the scope of > MODELLER but: > > PRES DISU -0.36 ! Patch for disulfides. Patch must be 1-CYS and 2-CYS. > > > ATOM 1:CB CT2 -0.10 ! > ATOM 1:SG SM -0.08 ! 2:SG--2:CB-- > ATOM 2:SG SM -0.08 ! / > ATOM 2:CB CT2 -0.10 ! -1:CB--1:SG > > So, I get the syntax but I fail to understand one thing. Is that number > -0.36 the overall charge of the residue or something?
Yes, that's the total charge. Modeller doesn't actually depend on this being correct, however, so you can set it to 0 if you like. You can't leave it blank though.
Ben Webb, Modeller Caretaker
Thanks for the answers Ben, they have been quite useful. In fact, they always are. Congrats to you and the team for the efforts.
So, I've managed to build the topology:
PRES THIO -0.18 ATOM 1:CB CT2 -0.10 ATOM 1:SG SM -0.08 ATOM 2:CD CC 0.55 ATOM 2:OE1 O -0.55 DELE ATOM 1:HG DELE ATOM 2:NE2 DELE ATOM 2:HE22 DELE ATOM 2:HE21 BOND 1:SG 2:CD
This is based on the DISU Topology in the FAQ. Since the concept is pretty much the same, I thought both bonds would share the same characteristics. The thioester is just a sulphur attached to a carbon, with the deletion of a NH2 group from the Glutamine and a H group from the Cysteine (in my case).
I then set this like:
self.patch(residue_type='THIO', residues=(self.residues['981:B'], self.residues['984:B']))
And it apparently works. My structure does show a bonding of the two residues, albeit quite crooked when compared with the original. This is perhaps, the rest of the protein, from the bond onward just disappears. In the visualisation program that is. In the PDB file, it all seems ok actually, but for the two lines next to the bond:
ATOM 7740 OE1 GLN B 984 -52.638 -33.311 -71.659 1.00326.32 O ATOM 7741 C GLN B 984 -51.735 -35.089 -72.840 1.00 0.00 C ATOM 7742 O GLN B 984 -51.735 -35.089 -72.840 1.00 0.00 O ATOM 7743 N ASN B 985 -50.555 -32.594 -69.198 1.00326.32 N
This is the carbonyl group of the Glutamine residue involved in the thioester linkage. The last two factors are completely mismatching the surroundings and I think I might have made a mistake when building the topology. Maybe missing a BOND statement.
I'm trying to solve it on my own, but I have no experience either with this, or with CHARMM and so I might be heading down the wrong path. That's why I'm posting here. If you can give me a hint on what I'm doing wrong, or anyone else, I'd be thankful.
Best regards to all,
João [ .. ] Rodrigues
(Blog) http://doeidoei.wordpress.com (MSN) always_asleep_@hotmail.com (Skype) rodrigues.jglm
João Rodrigues wrote: > So, I've managed to build the topology: ...
Everything looks OK to me, although you might want some IC (internal coordinate) entries in there too to get the initial conformation correct if you're going to build extended chains using this topology.
> In the visualisation program that is. In the PDB file, it all seems ok > actually, but for the two lines next to the bond: > > ATOM 7740 OE1 GLN B 984 -52.638 -33.311 -71.659 > 1.00326.32 O > ATOM 7741 C GLN B 984 -51.735 -35.089 -72.840 1.00 > 0.00 C > ATOM 7742 O GLN B 984 -51.735 -35.089 -72.840 1.00 > 0.00 O > ATOM 7743 N ASN B 985 -50.555 -32.594 -69.198 > 1.00326.32 N
I don't how that it would happen that two atoms would end up superposed like that, since you haven't changed the bonds within the GLN.
Ben Webb, Modeller Caretaker
Well, I actually dropped this patch and just substituted the GLN for a CYS and made a disulfide. Not what I wanted but it worked well enough ...
I'll try to work this out but honestly I have no clue what happened either. I'll try to make it work and if I have success I'll report back.
Best Regards and thanks a lot for the help.
João [ .. ] Rodrigues
(Blog) http://doeidoei.wordpress.com (MSN) always_asleep_@hotmail.com (Skype) rodrigues.jglm
participants (2)
-
João Rodrigues -
Modeller Caretaker