- Posts: 211
portein model stability
- Hegedus
-
Topic Author
- Visitor
9 years 4 months ago #4206
by Hegedus
portein model stability was created by Hegedus
Hi,
I would like to know if I can expect Martini to sign a wrong structural model of a membrane protein. E.g. if I have a low quality homology model generated, performing CG MD: should I see explosion, disassembling of the transmembrane domain ?
With all atom simulations it seems to be possible (see ref below), but atomistic simulations are resource expensive.
Molecular dynamics simulations and membrane protein structure quality.
Ivetac A, Sansom MS. www.ncbi.nlm.nih.gov/pubmed/17960373
Thanks for your help in advance,
Tamas
I would like to know if I can expect Martini to sign a wrong structural model of a membrane protein. E.g. if I have a low quality homology model generated, performing CG MD: should I see explosion, disassembling of the transmembrane domain ?
With all atom simulations it seems to be possible (see ref below), but atomistic simulations are resource expensive.
Molecular dynamics simulations and membrane protein structure quality.
Ivetac A, Sansom MS. www.ncbi.nlm.nih.gov/pubmed/17960373
Thanks for your help in advance,
Tamas
Please Log in or Create an account to join the conversation.
- Clement
-
- Offline
- Admin
Less
More
9 years 4 months ago #4207
by Clement
Replied by Clement on topic portein model stability
Hi Tamas,
Martini doesn't allow rearrangement of secondary structure; by that I mean the secondary structure is defined beforehand, and doesn't change during the simulation(s). If you have a poorly defined model, I assume your secondary structure won't be defined properly, and the following MD will not be physically relevant.
How costly would atomistic equilibrations would be? Maybe full description of solvent/membrane isn't required in a first step, you might be able to get a better model with enhanced sampling techniques... Good enough to be coarse-grained or so.
Good luck with that.
Martini doesn't allow rearrangement of secondary structure; by that I mean the secondary structure is defined beforehand, and doesn't change during the simulation(s). If you have a poorly defined model, I assume your secondary structure won't be defined properly, and the following MD will not be physically relevant.
How costly would atomistic equilibrations would be? Maybe full description of solvent/membrane isn't required in a first step, you might be able to get a better model with enhanced sampling techniques... Good enough to be coarse-grained or so.
Good luck with that.
Please Log in or Create an account to join the conversation.
- Hegedus
-
Topic Author
- Visitor
9 years 4 months ago #4208
by Hegedus
Replied by Hegedus on topic portein model stability
Hi,
I do not expect changes in sec structure because of CG ff properties.
However, if a charged helix is inside the membrane, I would expect significant changes in the tertiary structure compared to the initial structure. E.g. rearrangement of the helices or altered tilting of the helices. But everything seems to be relatively stable, RMSD is relatively small over the trajectory.
Atomistic equilibration is relatively fast. But you have to run equilibrium simulations to see conformational changes.
Thanks.
I do not expect changes in sec structure because of CG ff properties.
However, if a charged helix is inside the membrane, I would expect significant changes in the tertiary structure compared to the initial structure. E.g. rearrangement of the helices or altered tilting of the helices. But everything seems to be relatively stable, RMSD is relatively small over the trajectory.
Atomistic equilibration is relatively fast. But you have to run equilibrium simulations to see conformational changes.
Thanks.
Please Log in or Create an account to join the conversation.
- Clement
-
- Offline
- Admin
Less
More
- Posts: 211
9 years 4 months ago #4209
by Clement
Replied by Clement on topic portein model stability
Mmmmh... I'm not really sure of your question here. Sorry.
If a polar helix finds itself in a (apolar) moeity of a bilayer, it will want to move away; Martini is good enough to get that quite nicely actually. If you don't have elastic network on your protein (which is recommended, but probably not in this specific case of course), the tertiary structure will be affected by this.
Hoping I'm somehow replying to your question...
If a polar helix finds itself in a (apolar) moeity of a bilayer, it will want to move away; Martini is good enough to get that quite nicely actually. If you don't have elastic network on your protein (which is recommended, but probably not in this specific case of course), the tertiary structure will be affected by this.
Hoping I'm somehow replying to your question...
Please Log in or Create an account to join the conversation.
- Hegedus
-
Topic Author
- Visitor
9 years 4 months ago #4210
by Hegedus
Replied by Hegedus on topic portein model stability
Yes, you replied that I should see changes.
There was a wrong 3D model generated for my protein: 7 TM helices instead of 6 (from experiments we know that there is only 6TMH). I wanted to show that the quality of the 7 TM version could be also indicated by CG simulations. But it remained pretty close/similar to the starting structure (energy minimization in vacuum; insane.py; energy minimization; equilibration; production run - dt = 0.02, nsteps = 10000000, T=300; no elastic network, no constraints).
I will try higher temperature.
Moreover, it could happen that the wrong 7 TM is not so bad physically (the extra part might be enough hydrophobic - I should check).
There was a wrong 3D model generated for my protein: 7 TM helices instead of 6 (from experiments we know that there is only 6TMH). I wanted to show that the quality of the 7 TM version could be also indicated by CG simulations. But it remained pretty close/similar to the starting structure (energy minimization in vacuum; insane.py; energy minimization; equilibration; production run - dt = 0.02, nsteps = 10000000, T=300; no elastic network, no constraints).
I will try higher temperature.
Moreover, it could happen that the wrong 7 TM is not so bad physically (the extra part might be enough hydrophobic - I should check).
Please Log in or Create an account to join the conversation.
Time to create page: 0.105 seconds