Re: [gmx-users] Pulling inside a channel to calculate the PMF
Thanks Alex, finding the positions as you propose seems a legitimate way for this problem. One will have to wait enough with a classical MD but Bias MD should help. In my case the channel is not a large ion channel. FR. 2017-06-22 18:26 GMT+02:00 Alex : > I've never used PLUMED, to be honest, so I don't know. > > If ligands are salt ions or any other overall charged entities, I would > solvate the channel (with restraint, maybe) for essentially a production > run and apply an electric field "parallel" to the channel lumen, then just > visualize the trajectory and pick the configurations by hand. gmx trjconv > can -dump any particular frame of your liking. Maybe not as automated as > one could hope, but then you will know in-depth what configs you have. > > Alex > > > On 6/22/2017 3:03 AM, François-Régis Chalaoux wrote: > >> Hi Alex, >> >> There is no confusion about the pulling that is only a preparation for >> WHAM, but as you tell it you need to choose positions. These positions >> could be choosen manually: if one choose to explore the static channel >> from >> apo, then I'm agree, select the positions in the channel "in a bunch of >> reasonably spaced spots along the curved "axis" of your channel, each >> followed by equilibration and the production run, yielding the data useful >> for WHAM". As you say, you need time and patience. Could PLUMED (or >> somethoing else) /Gromacs be used to automate this ? >> >> On the other side, this channel come from a static APO structure and the >> reality of this channel is tiedous, so why not discover dynamically the >> channel and run the next equilibration and the production run on the fly, >> could >> PLUMED (or somethoing else)/Gromacs be used to automate this ? How to >> discover the channel on the fly ? May be it is feasible to use simply the >> coordinates of the channel discovered by MDPocket with a long classical MD >> realized previously on the apo/holo structure ? >> >> FR. >> >> 2017-06-22 10:19 GMT+02:00 Alex : >> >> I think your problem is solvable with enough perseverance, but there may >>> also be some confusion... The PMF calculation isn't based on the data >>> obtained from an actual pull: the pulling rate in those simulations is >>> set >>> to zero. The reason for using pull code there is to pull (pun intended) >>> the >>> positions and/or forces. The initial structures, of course, is where e.g. >>> Justin's tutorial uses pull along a straight line. >>> >>> However, if you have the time and patience, you could of course >>> (manually?) put your ligand in a bunch of reasonably spaced spots along >>> the >>> curved "axis" of your channel, each followed by equilibration and the >>> production run, yielding the data useful for WHAM. >>> >>> Hope this helps. >>> >>> Alex >>> >>> >>> On 6/22/2017 2:05 AM, François-Régis Chalaoux wrote: >>> >>> Hi everybody, I would like to pull a ligand in a channel from my protein and calculate the PMF with Umbrella sampling. This channel is probably not linear and thus have to be defined. Too much colisions appears along a simple one axis pulling and causing a catastrophic PMF calculation. Currently I'm working with Gromacs and I wondered if it is possible to define statically or dynamically the coordinates for the center of the channel and use it during the simulation (Pulling). One solution could be to define in the md_pull.mdp configuration file all the transition points of my pathway through which my ligand must pass from my APO structure determined with CAVER. Instantiation of Gromacs parameters of md_pull.mdp is not enough clear for me and PLUMED is may be a more direct solution. Currently I have no experience with PLUMED but It seems on paper to be able. An other choice would be to define dynamically the channel center, on-the-fly. PLUMED seems also capable to manage this sort of challenge but once again I have no idea of the feasibility of a such work. What do you think about this problem and of the tracks evoked above ? Cheers, FR. *Pulling inside a channel to calculate the PMF*. Available from: https://www.researchgate.net/post/Pulling_inside_a_channel_t o_calculate_the_PMF#594b792648954cf7f2759a48 [accessed Jun 22, 2017]. -- >>> Gromacs Users mailing list >>> >>> * Please search the archive at http://www.gromacs.org/Support >>> /Mailing_Lists/GMX-Users_List before posting! >>> >>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists >>> >>> * For (un)subscribe requests visit >>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or >>> send a mail to gmx-users-requ...@gromacs.org. >>> >> > -- > Gromacs Users mailing list > > * Please search the archive at http://www.gromacs.org/Support > /Mailing_Lists/GMX-Users_List before posting! > > * Can't post? Read http://www.gromacs.org/Support/Ma
Re: [gmx-users] Pulling inside a channel to calculate the PMF
I've never used PLUMED, to be honest, so I don't know. If ligands are salt ions or any other overall charged entities, I would solvate the channel (with restraint, maybe) for essentially a production run and apply an electric field "parallel" to the channel lumen, then just visualize the trajectory and pick the configurations by hand. gmx trjconv can -dump any particular frame of your liking. Maybe not as automated as one could hope, but then you will know in-depth what configs you have. Alex On 6/22/2017 3:03 AM, François-Régis Chalaoux wrote: Hi Alex, There is no confusion about the pulling that is only a preparation for WHAM, but as you tell it you need to choose positions. These positions could be choosen manually: if one choose to explore the static channel from apo, then I'm agree, select the positions in the channel "in a bunch of reasonably spaced spots along the curved "axis" of your channel, each followed by equilibration and the production run, yielding the data useful for WHAM". As you say, you need time and patience. Could PLUMED (or somethoing else) /Gromacs be used to automate this ? On the other side, this channel come from a static APO structure and the reality of this channel is tiedous, so why not discover dynamically the channel and run the next equilibration and the production run on the fly, could PLUMED (or somethoing else)/Gromacs be used to automate this ? How to discover the channel on the fly ? May be it is feasible to use simply the coordinates of the channel discovered by MDPocket with a long classical MD realized previously on the apo/holo structure ? FR. 2017-06-22 10:19 GMT+02:00 Alex : I think your problem is solvable with enough perseverance, but there may also be some confusion... The PMF calculation isn't based on the data obtained from an actual pull: the pulling rate in those simulations is set to zero. The reason for using pull code there is to pull (pun intended) the positions and/or forces. The initial structures, of course, is where e.g. Justin's tutorial uses pull along a straight line. However, if you have the time and patience, you could of course (manually?) put your ligand in a bunch of reasonably spaced spots along the curved "axis" of your channel, each followed by equilibration and the production run, yielding the data useful for WHAM. Hope this helps. Alex On 6/22/2017 2:05 AM, François-Régis Chalaoux wrote: Hi everybody, I would like to pull a ligand in a channel from my protein and calculate the PMF with Umbrella sampling. This channel is probably not linear and thus have to be defined. Too much colisions appears along a simple one axis pulling and causing a catastrophic PMF calculation. Currently I'm working with Gromacs and I wondered if it is possible to define statically or dynamically the coordinates for the center of the channel and use it during the simulation (Pulling). One solution could be to define in the md_pull.mdp configuration file all the transition points of my pathway through which my ligand must pass from my APO structure determined with CAVER. Instantiation of Gromacs parameters of md_pull.mdp is not enough clear for me and PLUMED is may be a more direct solution. Currently I have no experience with PLUMED but It seems on paper to be able. An other choice would be to define dynamically the channel center, on-the-fly. PLUMED seems also capable to manage this sort of challenge but once again I have no idea of the feasibility of a such work. What do you think about this problem and of the tracks evoked above ? Cheers, FR. *Pulling inside a channel to calculate the PMF*. Available from: https://www.researchgate.net/post/Pulling_inside_a_channel_t o_calculate_the_PMF#594b792648954cf7f2759a48 [accessed Jun 22, 2017]. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support /Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] Pulling inside a channel to calculate the PMF
Hi Alex, There is no confusion about the pulling that is only a preparation for WHAM, but as you tell it you need to choose positions. These positions could be choosen manually: if one choose to explore the static channel from apo, then I'm agree, select the positions in the channel "in a bunch of reasonably spaced spots along the curved "axis" of your channel, each followed by equilibration and the production run, yielding the data useful for WHAM". As you say, you need time and patience. Could PLUMED (or somethoing else) /Gromacs be used to automate this ? On the other side, this channel come from a static APO structure and the reality of this channel is tiedous, so why not discover dynamically the channel and run the next equilibration and the production run on the fly, could PLUMED (or somethoing else)/Gromacs be used to automate this ? How to discover the channel on the fly ? May be it is feasible to use simply the coordinates of the channel discovered by MDPocket with a long classical MD realized previously on the apo/holo structure ? FR. 2017-06-22 10:19 GMT+02:00 Alex : > I think your problem is solvable with enough perseverance, but there may > also be some confusion... The PMF calculation isn't based on the data > obtained from an actual pull: the pulling rate in those simulations is set > to zero. The reason for using pull code there is to pull (pun intended) the > positions and/or forces. The initial structures, of course, is where e.g. > Justin's tutorial uses pull along a straight line. > > However, if you have the time and patience, you could of course > (manually?) put your ligand in a bunch of reasonably spaced spots along the > curved "axis" of your channel, each followed by equilibration and the > production run, yielding the data useful for WHAM. > > Hope this helps. > > Alex > > > On 6/22/2017 2:05 AM, François-Régis Chalaoux wrote: > >> Hi everybody, >> >> I would like to pull a ligand in a channel from my protein and calculate >> the PMF with Umbrella sampling. This channel is probably not linear and >> thus have to be defined. Too much colisions appears along a simple one >> axis >> pulling and causing a catastrophic PMF calculation. >> >> Currently I'm working with Gromacs and I wondered if it is possible to >> define statically or dynamically the coordinates for the center of the >> channel and use it during the simulation (Pulling). >> >> One solution could be to define in the md_pull.mdp configuration file all >> the transition points of my pathway through which my ligand must pass from >> my APO structure determined with CAVER. >> Instantiation of Gromacs parameters of md_pull.mdp is not enough clear for >> me and PLUMED is may be a more direct solution. Currently I have no >> experience with PLUMED but It seems on paper to be able. >> >> An other choice would be to define dynamically the channel center, >> on-the-fly. PLUMED seems also capable to manage this sort of challenge but >> once again I have no idea of the feasibility of a such work. >> >> >> What do you think about this problem and of the tracks evoked above ? >> >> >> Cheers, FR. >> >> *Pulling inside a channel to calculate the PMF*. Available from: >> https://www.researchgate.net/post/Pulling_inside_a_channel_t >> o_calculate_the_PMF#594b792648954cf7f2759a48 >> [accessed Jun 22, 2017]. >> > > -- > Gromacs Users mailing list > > * Please search the archive at http://www.gromacs.org/Support > /Mailing_Lists/GMX-Users_List before posting! > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists > > * For (un)subscribe requests visit > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or > send a mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] Pulling inside a channel to calculate the PMF
I think your problem is solvable with enough perseverance, but there may also be some confusion... The PMF calculation isn't based on the data obtained from an actual pull: the pulling rate in those simulations is set to zero. The reason for using pull code there is to pull (pun intended) the positions and/or forces. The initial structures, of course, is where e.g. Justin's tutorial uses pull along a straight line. However, if you have the time and patience, you could of course (manually?) put your ligand in a bunch of reasonably spaced spots along the curved "axis" of your channel, each followed by equilibration and the production run, yielding the data useful for WHAM. Hope this helps. Alex On 6/22/2017 2:05 AM, François-Régis Chalaoux wrote: Hi everybody, I would like to pull a ligand in a channel from my protein and calculate the PMF with Umbrella sampling. This channel is probably not linear and thus have to be defined. Too much colisions appears along a simple one axis pulling and causing a catastrophic PMF calculation. Currently I'm working with Gromacs and I wondered if it is possible to define statically or dynamically the coordinates for the center of the channel and use it during the simulation (Pulling). One solution could be to define in the md_pull.mdp configuration file all the transition points of my pathway through which my ligand must pass from my APO structure determined with CAVER. Instantiation of Gromacs parameters of md_pull.mdp is not enough clear for me and PLUMED is may be a more direct solution. Currently I have no experience with PLUMED but It seems on paper to be able. An other choice would be to define dynamically the channel center, on-the-fly. PLUMED seems also capable to manage this sort of challenge but once again I have no idea of the feasibility of a such work. What do you think about this problem and of the tracks evoked above ? Cheers, FR. *Pulling inside a channel to calculate the PMF*. Available from: https://www.researchgate.net/post/Pulling_inside_a_channel_to_calculate_the_PMF#594b792648954cf7f2759a48 [accessed Jun 22, 2017]. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
