- Posts: 9
How I can make an EN using ElNeDyn for each chain of my protein?
- Melchor
-
Topic Author
- Offline
- Fresh Boarder
Less
More
12 years 5 months ago #841
by Melchor
How I can make an EN using ElNeDyn for each chain of my protein? was created by Melchor
Hi all,
I Have a protein with four chains and I want to make an Elastic Network for each of these chains.
First of all I want to know if there are any option available in the martinize.py file for do it. If it is not possible do it from the martinize.py file ( if is possible, I haven't see the correct option when i have read file), I have thought about possible solutions but I don't know if these solutions could be good choices or not.
1 - First, make a global EN using Elnedyn and then make a ndx file and remove the connections between different chains, ie only keep the connection between the beads of the same chain.
2 - Modify the original itp file ,that you obtain after using martinize.py, obtaining from this first one 4 files, one for each chain.
3 - From the original pdb file, do four different files , one for each chain, run the martinize.py for each pdb and then in the top file use the 4 itp files.
Whay do you think about it?
Best regards,
Melchor
I Have a protein with four chains and I want to make an Elastic Network for each of these chains.
First of all I want to know if there are any option available in the martinize.py file for do it. If it is not possible do it from the martinize.py file ( if is possible, I haven't see the correct option when i have read file), I have thought about possible solutions but I don't know if these solutions could be good choices or not.
1 - First, make a global EN using Elnedyn and then make a ndx file and remove the connections between different chains, ie only keep the connection between the beads of the same chain.
2 - Modify the original itp file ,that you obtain after using martinize.py, obtaining from this first one 4 files, one for each chain.
3 - From the original pdb file, do four different files , one for each chain, run the martinize.py for each pdb and then in the top file use the 4 itp files.
Whay do you think about it?
Best regards,
Melchor
Please Log in or Create an account to join the conversation.
- Clement
-
- Offline
- Admin
Less
More
- Posts: 211
12 years 5 months ago #842
by Clement
Replied by Clement on topic How I can make an EN using ElNeDyn for each chain of my protein?
By default, the martinize.py script creates a topology for each chain of your PDB file. If two or more chains are identical (homodimer), only one ITP file is outputted for the specific chain. If you decide to include an elastic network, it will be of course generated for each chain independently.
Please Log in or Create an account to join the conversation.
- Melchor
-
Topic Author
- Offline
- Fresh Boarder
Less
More
- Posts: 9
12 years 5 months ago #843
by Melchor
Replied by Melchor on topic How I can make an EN using ElNeDyn for each chain of my protein?
Ok.Thamk yopu for you answer. But i want to clarify something if could be possible.
I have 4 different chains, my protein is an heterodimer. When I make an Elastic Network using the option -elastic, I make a global Elastic Network and I only obtain one ITP file that is called Protein_A.itp.
When I use ElNeDyn, without the option -merge no, I obtain an itp file that is called ProteinA+ProteinB+ProteinC+ProteinD.itp. I believed that this itp is a global Elastic Network that recognizes different chains but make connections between these chains. That I want to have, is an independent elastic network for each chain, without connections between the different chains.
Maybe I am in an error, but I thought that with rhe Elnedyn I just have an Elastic Netwoek that connects all the chains.
Could you clarify me this point?
I have 4 different chains, my protein is an heterodimer. When I make an Elastic Network using the option -elastic, I make a global Elastic Network and I only obtain one ITP file that is called Protein_A.itp.
When I use ElNeDyn, without the option -merge no, I obtain an itp file that is called ProteinA+ProteinB+ProteinC+ProteinD.itp. I believed that this itp is a global Elastic Network that recognizes different chains but make connections between these chains. That I want to have, is an independent elastic network for each chain, without connections between the different chains.
Maybe I am in an error, but I thought that with rhe Elnedyn I just have an Elastic Netwoek that connects all the chains.
Could you clarify me this point?
Please Log in or Create an account to join the conversation.
- Clement
-
- Offline
- Admin
Less
More
- Posts: 211
12 years 5 months ago #844
by Clement
Replied by Clement on topic How I can make an EN using ElNeDyn for each chain of my protein?
Check your ITP, but if you did not use the -merge option (don't mention any -merge in your command line, I didn't test it but I doubt "-merge no" is a good idea), each chain of your protein complex is described independently in your ITP file (you'll have four entries in your case, from "Protein_A" to "Protein_D", each entries containing "[atoms]", "[bonds]", etc. tags). As well as the elastic networks, defined for each chain independently from the others, i.e. without bonds linking them.
Please Log in or Create an account to join the conversation.
- Melchor
-
Topic Author
- Offline
- Fresh Boarder
Less
More
- Posts: 9
12 years 4 months ago #851
by Melchor
Replied by Melchor on topic How I can make an EN using ElNeDyn for each chain of my protein?
Hi,
Finally the discussion that I had started days ago, finished when I speaked with Clement via email about possible solutions to my homotetramer protein.
These solutions were:
Use the -sep and -merge no options or delete from the original pdb all the chains except one and then put in the itp file, of the whole system, four times the itp file of the chain, because is an homotetramer. Both solutions have worked properly.
Now I have another related problem. I have other homotetramer protein, 9LDT. I want to run a cg dynamics using enledyn with -ef 500 -eu 0.75, in the same way that for my other protein.
Well, in this case I can run the simulation if I don't use the -elnedyn option, but when I try to create the elastic network, the simulation collapsed after few steps due to a LINCS error. I have tried to run simulations with only one monomer, changing the time step, the nstlist and the rlist, changing the size of the box, changing the shape of the box, and I can't run the simulation. Could some one give me any idea of what is happening or what I can do to run the dynamics properly?
Finally the discussion that I had started days ago, finished when I speaked with Clement via email about possible solutions to my homotetramer protein.
These solutions were:
Use the -sep and -merge no options or delete from the original pdb all the chains except one and then put in the itp file, of the whole system, four times the itp file of the chain, because is an homotetramer. Both solutions have worked properly.
Now I have another related problem. I have other homotetramer protein, 9LDT. I want to run a cg dynamics using enledyn with -ef 500 -eu 0.75, in the same way that for my other protein.
Well, in this case I can run the simulation if I don't use the -elnedyn option, but when I try to create the elastic network, the simulation collapsed after few steps due to a LINCS error. I have tried to run simulations with only one monomer, changing the time step, the nstlist and the rlist, changing the size of the box, changing the shape of the box, and I can't run the simulation. Could some one give me any idea of what is happening or what I can do to run the dynamics properly?
Please Log in or Create an account to join the conversation.
- Clement
-
- Offline
- Admin
Less
More
- Posts: 211
12 years 4 months ago #852
by Clement
10 steps of minimization with any AA forcefield proposed by the pdb2gmx command will do the trick. Then you coarse-grained it...
I think the problem comes from what I mentioned above.
Good luck, keep us informed!
Replied by Clement on topic How I can make an EN using ElNeDyn for each chain of my protein?
Well, in this case I can run the simulation if I don't use the -elnedyn option, but when I try to create the elastic network, the simulation collapsed after few steps due to a LINCS error.
I think you need to minimize your AA structure first. The length of the bonds of the elastic network are taken from the initial set of coordinates, which are most of the time not optimized.10 steps of minimization with any AA forcefield proposed by the pdb2gmx command will do the trick. Then you coarse-grained it...
I have tried to run simulations with only one monomer, changing the time step, the nstlist and the rlist, changing the size of the box, changing the shape of the box, and I can't run the simulation.
Wow wow. Don't play too much with these options (rlist/ntslit), they are more or less forcefield dependent. And if it doesn't work with one size/shape of box, it has many chances not to work with different ones...I think the problem comes from what I mentioned above.
Good luck, keep us informed!
Please Log in or Create an account to join the conversation.
- Melchor
-
Topic Author
- Offline
- Fresh Boarder
Less
More
- Posts: 9
12 years 4 months ago #853
by Melchor
Replied by Melchor on topic How I can make an EN using ElNeDyn for each chain of my protein?
Hi,
Unfortunately Clement, this does ot solve my problem.
I run AA minimization of 10.20,200 even 2000 steps with the Gromos force field, but then doing the CG happens the same when I try to make the Elastic Network.
I follow trying to equilibrate my system , running more minimizations and so on, but until now I don't have any succes.
I have to note that I don't use the 9LDT that you can download from the PDB I use a mutant of this protein, maybe the structure is not very suitable, but with this structure, I can run CG dynamics, if I do not use elnedyn, and also AA dynamics of hundreds of nanoseconds.
For these reasons, I don't know what is happening with my system
Unfortunately Clement, this does ot solve my problem.
I run AA minimization of 10.20,200 even 2000 steps with the Gromos force field, but then doing the CG happens the same when I try to make the Elastic Network.
I follow trying to equilibrate my system , running more minimizations and so on, but until now I don't have any succes.
I have to note that I don't use the 9LDT that you can download from the PDB I use a mutant of this protein, maybe the structure is not very suitable, but with this structure, I can run CG dynamics, if I do not use elnedyn, and also AA dynamics of hundreds of nanoseconds.
For these reasons, I don't know what is happening with my system
Please Log in or Create an account to join the conversation.
- Melchor
-
Topic Author
- Offline
- Fresh Boarder
Less
More
- Posts: 9
12 years 4 months ago #854
by Melchor
Replied by Melchor on topic How I can make an EN using ElNeDyn for each chain of my protein?
Hi,
I can add more information about my problem.
I have discovered that if use -elastic option and also -elastic + -sep , I can run a CG dynamics properly.
The problem is when I try to use -elnedyn.
I don't know if this data could clarify more my problem for you, but I am more or less at the same point because I want to use elnedyn.
I can add more information about my problem.
I have discovered that if use -elastic option and also -elastic + -sep , I can run a CG dynamics properly.
The problem is when I try to use -elnedyn.
I don't know if this data could clarify more my problem for you, but I am more or less at the same point because I want to use elnedyn.
Please Log in or Create an account to join the conversation.
- Clement
-
- Offline
- Admin
Less
More
- Posts: 211
12 years 4 months ago #855
by Clement
Replied by Clement on topic How I can make an EN using ElNeDyn for each chain of my protein?
Thanks for that update. We're gonna have a look at that asap. There's definitely a problem with the martinize.py script.
In the meantime, can't you just create the ElNeDyn network on one monomer, then use it for all the other? It will avoid the -sep option, which seems to be problematic with ElNeDyn.
I'll keep you informed.
In the meantime, can't you just create the ElNeDyn network on one monomer, then use it for all the other? It will avoid the -sep option, which seems to be problematic with ElNeDyn.
I'll keep you informed.
Please Log in or Create an account to join the conversation.
- Melchor
-
Topic Author
- Offline
- Fresh Boarder
Less
More
- Posts: 9
12 years 4 months ago #856
by Melchor
Replied by Melchor on topic How I can make an EN using ElNeDyn for each chain of my protein?
Hi Clement,
Thank you for answer.
I can't do it because in fact is my problem. Because before, with the first system, the -sep did not work for and I can't separate the chains, for this I make the "strategy" of split the pdb and take a monomer , create the topology the itp file from it and the use 4 times this topology.
But with this new system , I can't do it because always that I try to use -elnedyn, when I try to run the dynamics after, at some point my simulation stops because the simulation collapsed.
Thank you for answer.
I can't do it because in fact is my problem. Because before, with the first system, the -sep did not work for and I can't separate the chains, for this I make the "strategy" of split the pdb and take a monomer , create the topology the itp file from it and the use 4 times this topology.
But with this new system , I can't do it because always that I try to use -elnedyn, when I try to run the dynamics after, at some point my simulation stops because the simulation collapsed.
Please Log in or Create an account to join the conversation.
Time to create page: 0.103 seconds