- Posts: 416
Particles that act as repulsive spheres
- shaharsu
-
Topic Author
- Visitor
12 years 9 months ago - 12 years 9 months ago #742
by shaharsu
Particles that act as repulsive spheres was created by shaharsu
Hi everyone,
I was wondering if it is possible to model a repulsive sphere in MARTINI, and what would be the parameters. I'm interested in inserting spheres of a rather large radius (roughly 2 nm) that would be repulsive to proteins but would be easily dissolved in water. Alternatively - they can just be purely repulsive, and thus avoid self-aggregation. From what I've gathered the only way to modify the radius is using different C6 and C12 terms - but that would also modify the strength of interaction...
Would appreciate any insight you might have on this problem.
Shahar.
I was wondering if it is possible to model a repulsive sphere in MARTINI, and what would be the parameters. I'm interested in inserting spheres of a rather large radius (roughly 2 nm) that would be repulsive to proteins but would be easily dissolved in water. Alternatively - they can just be purely repulsive, and thus avoid self-aggregation. From what I've gathered the only way to modify the radius is using different C6 and C12 terms - but that would also modify the strength of interaction...
Would appreciate any insight you might have on this problem.
Shahar.
Please Log in or Create an account to join the conversation.
- xavier
-
- Offline
- Admin
Less
More
12 years 9 months ago #745
by xavier
Replied by xavier on topic Particles that act as repulsive spheres
Dear Shahar,
It is not entirely clear what you like to do.
Creating a "particle" that would have a 2 nm radius is possible. The radius is defined by the LJ potential, roughly by the minimum of the potential. So you could modify the C6/C12 to obtain a particle that have such minimum, no problem.
Then it is not clear how you'd like this particle to interact with itself and how you actually define self-aggregation ... self-aggregation of what?
You need to give more details for us to help you.
XAvier.
It is not entirely clear what you like to do.
Creating a "particle" that would have a 2 nm radius is possible. The radius is defined by the LJ potential, roughly by the minimum of the potential. So you could modify the C6/C12 to obtain a particle that have such minimum, no problem.
Then it is not clear how you'd like this particle to interact with itself and how you actually define self-aggregation ... self-aggregation of what?
You need to give more details for us to help you.
XAvier.
shaharsu wrote: Hi everyone,
I was wondering if it is possible to model a repulsive sphere in MARTINI, and what would be the parameters. I'm interested in inserting spheres of a rather large radius (roughly 2 nm) that would be repulsive to proteins but would be easily dissolved in water. Alternatively - they can just be purely repulsive, and thus avoid self-aggregation. From what I've gathered the only way to modify the radius is using different C6 and C12 terms - but that would also modify the strength of interaction...
Would appreciate any insight you might have on this problem.
Shahar.
Please Log in or Create an account to join the conversation.
- shaharsu
-
Topic Author
- Visitor
12 years 9 months ago #746
by shaharsu
Replied by shaharsu on topic Particles that act as repulsive spheres
Hi Xavier,
Thanks very much for the quick reply! My intention is for the spheres to be completely repulsive for all interactions - both with itself and the other molecules in the simulation.
Shahar.
Thanks very much for the quick reply! My intention is for the spheres to be completely repulsive for all interactions - both with itself and the other molecules in the simulation.
Shahar.
Please Log in or Create an account to join the conversation.
- shaharsu
-
Topic Author
- Visitor
12 years 9 months ago #748
by shaharsu
Replied by shaharsu on topic Particles that act as repulsive spheres
Hi again,
To make my questions a bit more clear -
My box should contain water, peptides and these particles. I want the particles to have a radius of 2.0 nm and an epsilon as close as possible to 0 - meaning the particles will be purely repulsive (hard spheres). I wanted to know how to achieve this.
From what I understand I can calculate eps and sigma with the following relations:
sigma=(c12/c6)^(1/6)
epsilon=c6^2 / (4*c12)
I then need to calculate sigma for each particle pair so that sigma = 2.0 + r and find the appropriate epsilon. I have a few questions:
1. Is this the correct methodology? or is there some other way to achieve this?
2. How low do I need my epsilon to be? In the .itp file it says that "super repulsive" interactions (type IX) have an epsilon of 2.0 - but this still implies some attraction!
3. In order to do this I will probably need to insert c6 and c12 values that are larger than 1 - is this problematic?
Thanks in advance!
Shahar.
To make my questions a bit more clear -
My box should contain water, peptides and these particles. I want the particles to have a radius of 2.0 nm and an epsilon as close as possible to 0 - meaning the particles will be purely repulsive (hard spheres). I wanted to know how to achieve this.
From what I understand I can calculate eps and sigma with the following relations:
sigma=(c12/c6)^(1/6)
epsilon=c6^2 / (4*c12)
I then need to calculate sigma for each particle pair so that sigma = 2.0 + r and find the appropriate epsilon. I have a few questions:
1. Is this the correct methodology? or is there some other way to achieve this?
2. How low do I need my epsilon to be? In the .itp file it says that "super repulsive" interactions (type IX) have an epsilon of 2.0 - but this still implies some attraction!
3. In order to do this I will probably need to insert c6 and c12 values that are larger than 1 - is this problematic?
Thanks in advance!
Shahar.
Please Log in or Create an account to join the conversation.
- djurre
-
- Offline
- Admin
Less
More
- Posts: 272
12 years 9 months ago #749
by djurre
Replied by djurre on topic Particles that act as repulsive spheres
Hi Shahar,
The 'Super repulsive' interaction in Martini is repulsive relative to all other interactions.
Every particle is purely repulsive: the (LJ)potential approaches infinity at small distances. This simply defines the size of the particle. If you would make particle repulsive up to large distance (eg by using a tabulated potential), this would make the particle bigger.
If a particle has a 'Super repulsive' interaction with everything (including itself), it will probably start to cluster with it's own 'alikes'.
Also, keep in mind, if you change the particle size to 2nm (really big!) it is strictly speaking no longer Martini and the behaviour of the rest of the system might not make sense at all.
If that still is what you want, you should probably use tabulated potentials to get a positive 'attraction'. See the Gromacs manual how to use them.
Groetnis,
Djurre
The 'Super repulsive' interaction in Martini is repulsive relative to all other interactions.
Every particle is purely repulsive: the (LJ)potential approaches infinity at small distances. This simply defines the size of the particle. If you would make particle repulsive up to large distance (eg by using a tabulated potential), this would make the particle bigger.
If a particle has a 'Super repulsive' interaction with everything (including itself), it will probably start to cluster with it's own 'alikes'.
Also, keep in mind, if you change the particle size to 2nm (really big!) it is strictly speaking no longer Martini and the behaviour of the rest of the system might not make sense at all.
If that still is what you want, you should probably use tabulated potentials to get a positive 'attraction'. See the Gromacs manual how to use them.
Groetnis,
Djurre
Please Log in or Create an account to join the conversation.
- xavier
-
- Offline
- Admin
Less
More
- Posts: 416
12 years 9 months ago #751
by xavier
Replied by xavier on topic Particles that act as repulsive spheres
Dear Shaharsu,
It is not completely clear what you exactly want to mimic with these particle but if you want to add particle that are repulsive to everything in your system even themselves it is not sure what they are going to do at the end ...
As mentioned Djurre they might aggregate with the like-particle but this is going to be completely defined by the relative strengths of the interactions (repulsions) that you will defined. This won't be easy to parametrize!
In principle you can use a purely repulsive term defined but a C12 parameter. What you'll need to do is to tune the strength of the repulsion of the self-interactions, the solvent and the protein particle(s). These combined with the other interactions, the pressure/density will result in a mixture of the particles. Then it is your task to decide which parameters will reproduce the behavior you are looking for ... probably the most difficult part!
The radius r=2.0 nm of your big particles will be the result of the combination of all the interactions in your system. It is an effective radius.
XAvier.
It is not completely clear what you exactly want to mimic with these particle but if you want to add particle that are repulsive to everything in your system even themselves it is not sure what they are going to do at the end ...
As mentioned Djurre they might aggregate with the like-particle but this is going to be completely defined by the relative strengths of the interactions (repulsions) that you will defined. This won't be easy to parametrize!
In principle you can use a purely repulsive term defined but a C12 parameter. What you'll need to do is to tune the strength of the repulsion of the self-interactions, the solvent and the protein particle(s). These combined with the other interactions, the pressure/density will result in a mixture of the particles. Then it is your task to decide which parameters will reproduce the behavior you are looking for ... probably the most difficult part!
The radius r=2.0 nm of your big particles will be the result of the combination of all the interactions in your system. It is an effective radius.
XAvier.
shaharsu wrote: Hi again,
To make my questions a bit more clear -
My box should contain water, peptides and these particles. I want the particles to have a radius of 2.0 nm and an epsilon as close as possible to 0 - meaning the particles will be purely repulsive (hard spheres). I wanted to know how to achieve this.
From what I understand I can calculate eps and sigma with the following relations:
sigma=(c12/c6)^(1/6)
epsilon=c6^2 / (4*c12)
I then need to calculate sigma for each particle pair so that sigma = 2.0 + r and find the appropriate epsilon. I have a few questions:
1. Is this the correct methodology? or is there some other way to achieve this?
2. How low do I need my epsilon to be? In the .itp file it says that "super repulsive" interactions (type IX) have an epsilon of 2.0 - but this still implies some attraction!
3. In order to do this I will probably need to insert c6 and c12 values that are larger than 1 - is this problematic?
Thanks in advance!
Shahar.
Please Log in or Create an account to join the conversation.
- shaharsu
-
Topic Author
- Visitor
12 years 9 months ago #752
by shaharsu
Replied by shaharsu on topic Particles that act as repulsive spheres
Hi Xavier and Djurre,
Again, thanks very much for your helpful comments! It's very encouraging to work with Martini when there's so much response and assitance from the developers!
Some insight into what I'm trying to do: The hard spheres are supposed to model non-interacting, macromolecular crowders in my simulation box. Therefore there is a sound (and neccessary!) physical basis for their "non-interaction". I'm trying to see how my system is affected by these creatures taking up space in the simulation box.
I've tried your suggestions: as to dealing with the martini particle C6 and C12 terms - this is (as you've hinted :) ) a major headache! even when the epsilon for the self interaction LJ term is 2 orders of magnitude smaller than for cross interactions I still get aggregation of these large particles - no idea why this happens. Any ideas?
As for the tabulated potentials - this seems like a promising approach but I have some questions regarding the creation of the tables. From my understanding I cannot specify that I want to use a user defined function together with a shifted potential. Do I need to manually adjust my table.xvg to decay to 0? If so - should I use the equation shown in the martini FAQ (the one for phi(r)) to define the value for the potential?
Thanks again!
Shahar.
Again, thanks very much for your helpful comments! It's very encouraging to work with Martini when there's so much response and assitance from the developers!
Some insight into what I'm trying to do: The hard spheres are supposed to model non-interacting, macromolecular crowders in my simulation box. Therefore there is a sound (and neccessary!) physical basis for their "non-interaction". I'm trying to see how my system is affected by these creatures taking up space in the simulation box.
I've tried your suggestions: as to dealing with the martini particle C6 and C12 terms - this is (as you've hinted :) ) a major headache! even when the epsilon for the self interaction LJ term is 2 orders of magnitude smaller than for cross interactions I still get aggregation of these large particles - no idea why this happens. Any ideas?
As for the tabulated potentials - this seems like a promising approach but I have some questions regarding the creation of the tables. From my understanding I cannot specify that I want to use a user defined function together with a shifted potential. Do I need to manually adjust my table.xvg to decay to 0? If so - should I use the equation shown in the martini FAQ (the one for phi(r)) to define the value for the potential?
Thanks again!
Shahar.
xavier wrote: Dear Shaharsu,
It is not completely clear what you exactly want to mimic with these particle but if you want to add particle that are repulsive to everything in your system even themselves it is not sure what they are going to do at the end ...
As mentioned Djurre they might aggregate with the like-particle but this is going to be completely defined by the relative strengths of the interactions (repulsions) that you will defined. This won't be easy to parametrize!
In principle you can use a purely repulsive term defined but a C12 parameter. What you'll need to do is to tune the strength of the repulsion of the self-interactions, the solvent and the protein particle(s). These combined with the other interactions, the pressure/density will result in a mixture of the particles. Then it is your task to decide which parameters will reproduce the behavior you are looking for ... probably the most difficult part!
The radius r=2.0 nm of your big particles will be the result of the combination of all the interactions in your system. It is an effective radius.
XAvier.
shaharsu wrote: Hi again,
To make my questions a bit more clear -
My box should contain water, peptides and these particles. I want the particles to have a radius of 2.0 nm and an epsilon as close as possible to 0 - meaning the particles will be purely repulsive (hard spheres). I wanted to know how to achieve this.
From what I understand I can calculate eps and sigma with the following relations:
sigma=(c12/c6)^(1/6)
epsilon=c6^2 / (4*c12)
I then need to calculate sigma for each particle pair so that sigma = 2.0 + r and find the appropriate epsilon. I have a few questions:
1. Is this the correct methodology? or is there some other way to achieve this?
2. How low do I need my epsilon to be? In the .itp file it says that "super repulsive" interactions (type IX) have an epsilon of 2.0 - but this still implies some attraction!
3. In order to do this I will probably need to insert c6 and c12 values that are larger than 1 - is this problematic?
Thanks in advance!
Shahar.
Please Log in or Create an account to join the conversation.
- xavier
-
- Offline
- Admin
Less
More
- Posts: 416
12 years 9 months ago #755
by xavier
Replied by xavier on topic Particles that act as repulsive spheres
Dear Shahar,
It is always good that people find us useful :))
The reason for the big particle to collapse is not clear as it might result to many things. It however most likely relates to obvious physical behavior of particle mixtures ... but I am not an expert here. Entropy?
Is it a problem that they interact? What evidence you have that they do not interact beside their name?
What mass did you give to your big particle? It should probably be much larger than the regular one used for a Martini particle. That will affect their diffusion and might impact on their overall behavior.
For the tabulated potential I believe the shift/switch function are applies to the potential you give. You can I believe use the debbug option to print out the potential you give as tabulated ... check out the GROMACS manual.
I hope this helps,
XAvier.
It is always good that people find us useful :))
The reason for the big particle to collapse is not clear as it might result to many things. It however most likely relates to obvious physical behavior of particle mixtures ... but I am not an expert here. Entropy?
Is it a problem that they interact? What evidence you have that they do not interact beside their name?
What mass did you give to your big particle? It should probably be much larger than the regular one used for a Martini particle. That will affect their diffusion and might impact on their overall behavior.
For the tabulated potential I believe the shift/switch function are applies to the potential you give. You can I believe use the debbug option to print out the potential you give as tabulated ... check out the GROMACS manual.
I hope this helps,
XAvier.
shaharsu wrote: Hi Xavier and Djurre,
Again, thanks very much for your helpful comments! It's very encouraging to work with Martini when there's so much response and assitance from the developers!
Some insight into what I'm trying to do: The hard spheres are supposed to model non-interacting, macromolecular crowders in my simulation box. Therefore there is a sound (and neccessary!) physical basis for their "non-interaction". I'm trying to see how my system is affected by these creatures taking up space in the simulation box.
I've tried your suggestions: as to dealing with the martini particle C6 and C12 terms - this is (as you've hinted :) ) a major headache! even when the epsilon for the self interaction LJ term is 2 orders of magnitude smaller than for cross interactions I still get aggregation of these large particles - no idea why this happens. Any ideas?
As for the tabulated potentials - this seems like a promising approach but I have some questions regarding the creation of the tables. From my understanding I cannot specify that I want to use a user defined function together with a shifted potential. Do I need to manually adjust my table.xvg to decay to 0? If so - should I use the equation shown in the martini FAQ (the one for phi(r)) to define the value for the potential?
Thanks again!
Shahar.
Please Log in or Create an account to join the conversation.
Time to create page: 0.108 seconds