Re: [gmx-users] Simulating crystalls
Does this give any clue. It shows large PE, KE and temp. Could that be a reason? More information required? However, I have chenged my system a bit. My conf.gro file has just 2 atoms. one at the origin and the other at the body center. I use genconf to create a mesh of 4x4x4 unit cells at a distance of 0 0 0. I thought this is a better way to simulate a set of crystals. md.log file: Max number of graph edges per atom is 0 Table routines are used for coulomb: TRUE Table routines are used for vdw: TRUE Constants for short-range and fourier stuff: r1 = 0.000, rc = 0.500 A = -8.000e+01, B = 1.280e+02, C = 3.333e+00, FourPi_V = 8.306e+00 c1 = -1.000e+01, c2 = 3.000e+01, c3 = -1.250e+00 c4 = -5.000e+00, c5 = -3.000e+01, c6 = 0.000e+00, N0 = 9.817e-02 Using shifted Lennard-Jones, switch between 0 and 0.5 nm Cut-off's: NS: 0.4 Coulomb: 0.5 LJ: 0.5 Generated table with 750 data points for Shift. Tabscale = 500 points/nm Generated table with 750 data points for LJ6Shift. Tabscale = 500 points/nm Generated table with 750 data points for LJ12Shift. Tabscale = 500 points/nm Going to determine what solvent types we have. There are 128 molecules, 128 charge groups and 128 atoms There are 0 optimized solvent molecules on node 0 There are 0 optimized water molecules on node 0 Center of mass motion removal mode is Linear We have the following groups for center of mass motion removal: 0: rest, initial mass: 7148.42 There are: 128 Atom Removing pbc first time Done rmpbc Started mdrun on node 0 Fri May 26 10:15:40 2006 Initial temperature: 272.815 K Step Time Lambda 0 0.0 0.0 Grid: 4 x 4 x 4 cells calc_bor: cg0=0, cg1=128, ncg=128 CG0[0]=0, CG1[0]=128 CG0[1]=0, CG1[1]=0 calc_bor: cg0=0, cg1=128, ncg=128 CG0[0]=0, CG1[0]=128 CG0[1]=0, CG1[1]=0 There are 128 atoms in your xtc output selectionLarge VCM(group rest): 0.0, -0.0, -0.0, ekin-cm: 8.92704e-17 Long Range LJ corr. to Epot: 0, Pres: 0, Vir: 0 Energies (kJ/mol) LJ (SR) LJ (LR) Disper. corr. Coulomb (SR) Coulomb (LR) 3.98375e+07 6.17853e+04 0.0e+00 0.0e+00 0.0e+00 Potential Kinetic En. Total Energy Temperature Pressure (bar) 3.98992e+07 3.59069e+04 3.99352e+07 2.26697e+04 1.79321e+09 Total Virial (kJ/mol) -8.16795e+07 -1.31844e+04 -1.31872e+04 -1.31832e+04 -8.16795e+07 -1.31882e+04 -1.31904e+04 -1.31864e+04 -8.16795e+07 Pressure (bar) 1.79321e+09 5.53395e+05 5.44391e+05 5.53370e+05 1.79322e+09 5.49020e+05 5.44463e+05 5.48982e+05 1.79320e+09 Total Dipole (Debye) 0.0e+00 0.0e+00 0.0e+00 Large VCM(group rest): 0.0, -0.0, -0.0, ekin-cm: 1.41064e-16 Large VCM(group rest): -0.2, 0.0, 0.1, ekin-cm: 1.95671e-06 -On 5/25/06, Mark Abraham [EMAIL PROTECTED] wrote: karamyog singh wrote: I am trying to simulate a crystal too. I have a box size of .65 nm and there are 8 bcc crystals in the system. Lattice parameter = 0.287 nm. I get a segmentation fault after mdrun -v. Is an atom with co-ordinates 0.574 0.000 0.000 equivalent to 0.287 0.000 0.000 in such a system?Cubic PBC with side length l means that (x,y,z) is equivalent to(x+al,y+bl,z+cl) for any integers a,b and c. What I am trying to ask is, does GROMACS consider these as 2 different atoms placed at the respective positions or do the 2 atoms overlap?No. GROMACS takes every atom you give in the .top file and uses it -there is no attempt to check for symmetry simplifications such asquantum chemistry codes do. What could be the reason for the segmentation fault? Is it because the atoms are too close to each other?You're on your own here unless you can provide more information.Mark___ gmx-users mailing listgmx-users@gromacs.orghttp://www.gromacs.org/mailman/listinfo/gmx-usersPlease don't post (un)subscribe requests to the list. Use the www interface or send it to [EMAIL PROTECTED].Can't post? Read http://www.gromacs.org/mailing_lists/users.php ___ gmx-users mailing listgmx-users@gromacs.org http://www.gromacs.org/mailman/listinfo/gmx-users Please don't post (un)subscribe requests to the list. Use the www interface or send it to [EMAIL PROTECTED] Can't post? Read http://www.gromacs.org/mailing_lists/users.php
Re: [gmx-users] Simulating crystalls
karamyog singh wrote: Does this give any clue. It shows large PE, KE and temp. Could that be a reason? More information required? Your PE is ridiculously large. You have order 10^5 kJ/mol of PE for every atom, and obviously that is only coming from your LJ function. Check how you generated that and told gromacs what it should be. Be careful of units and multiples of 1000! In the absence of temperature control, that PE will equipartition to the KE degrees of freedom, leading to the other observations of high KE and temperature. However, I have chenged my system a bit. My conf.gro file has just 2 atoms. one at the origin and the other at the body center. I use genconf to create a mesh of 4x4x4 unit cells at a distance of 0 0 0. I thought this is a better way to simulate a set of crystals. Sounds like a better way to generate your simulation cells. You shouldn't now have overlapping atoms (check the output .gro file yourself) but you will have a simulation cell that is initially symmetric, until the gen_vel random KE breaks it. Mark ___ gmx-users mailing listgmx-users@gromacs.org http://www.gromacs.org/mailman/listinfo/gmx-users Please don't post (un)subscribe requests to the list. Use the www interface or send it to [EMAIL PROTECTED] Can't post? Read http://www.gromacs.org/mailing_lists/users.php
Re: [gmx-users] Simulating crystalls
However if the gen_velocity breaks up my system and at the end of my simulation, I do not get a bcc structure at lets say 273 K, which I am not getting, then how can I consider my simulation to be correct? The structure should remain the same. Or could it be that since my structure is a nano layer film, therefore it is unstable and explodes and I should not compare my results with a bulk material? Moreover, to avoid the error in the LJ, I have changed my system to copper and have used the potential parameters given in gromacs. Temperature of simulation is 273 and I have used the option no for Pressure coupling because i want my system to be at 1 bar. - karamyog.On 5/29/06, Mark Abraham [EMAIL PROTECTED] wrote: karamyog singh wrote: Does this give any clue. It shows large PE, KE and temp. Could that be a reason? More information required?Your PE is ridiculously large. You have order 10^5 kJ/mol of PE for every atom, and obviously that is only coming from your LJ function.Check how you generated that and told gromacs what it should be. Becareful of units and multiples of 1000! In the absence of temperaturecontrol, that PE will equipartition to the KE degrees of freedom, leading to the other observations of high KE and temperature. However, I have chenged my system a bit. My conf.gro file has just 2 atoms. one at the origin and the other at the body center. I use genconf to create a mesh of 4x4x4 unit cells at a distance of 0 0 0.I thought this is a better way to simulate a set of crystals.Sounds like a better way to generate your simulation cells. Youshouldn't now have overlapping atoms (check the output .gro file yourself) but you will have a simulation cell that is initiallysymmetric, until the gen_vel random KE breaks it.Mark___gmx-users mailing list gmx-users@gromacs.orghttp://www.gromacs.org/mailman/listinfo/gmx-usersPlease don't post (un)subscribe requests to the list. Use thewww interface or send it to [EMAIL PROTECTED].Can't post? Read http://www.gromacs.org/mailing_lists/users.php ___ gmx-users mailing listgmx-users@gromacs.org http://www.gromacs.org/mailman/listinfo/gmx-users Please don't post (un)subscribe requests to the list. Use the www interface or send it to [EMAIL PROTECTED] Can't post? Read http://www.gromacs.org/mailing_lists/users.php
Re: [gmx-users] Simulating crystalls
karamyog singh wrote: However if the gen_velocity breaks up my system and at the end of my simulation, I do not get a bcc structure at lets say 273 K, which I am not getting, then how can I consider my simulation to be correct? The structure should remain the same. I don't get the impression you read my post. I don't have to spend my time helping you :-) I think gen_velocity is not breaking the system, your LJ potential is breaking the system because energy is flowing from PE to KE. See my post below again. Or could it be that since my structure is a nano layer film, therefore it is unstable and explodes and I should not compare my results with a bulk material? I wouldn't know. Moreover, to avoid the error in the LJ, I have changed my system to copper and have used the potential parameters given in gromacs. Temperature of simulation is 273 and I have used the option no for Pressure coupling because i want my system to be at 1 bar. That just gives you NVT or NVE, not NPT or NPE with P=1 bar. You would only have a pressure of 1 bar there if you did so by construction. Mark karamyog singh wrote: Does this give any clue. It shows large PE, KE and temp. Could that be a reason? More information required? Your PE is ridiculously large. You have order 10^5 kJ/mol of PE for every atom, and obviously that is only coming from your LJ function. Check how you generated that and told gromacs what it should be. Be careful of units and multiples of 1000! In the absence of temperature control, that PE will equipartition to the KE degrees of freedom, leading to the other observations of high KE and temperature. ___ gmx-users mailing listgmx-users@gromacs.org http://www.gromacs.org/mailman/listinfo/gmx-users Please don't post (un)subscribe requests to the list. Use the www interface or send it to [EMAIL PROTECTED] Can't post? Read http://www.gromacs.org/mailing_lists/users.php
Re: [gmx-users] Simulating crystalls
You're on your own here unless you can provide more information. here is my .mdp file too. - karamyog. title = Yo cpp = /usr/bin/cpp include = define = integrator = md tinit = 0 dt = 0.01 nsteps = 200 init_step = 0 comm-mode = Linear nstcomm = 1 bd-temp = 273 bd-fric = 0 ld-seed = 1993 emtol = 100 emstep = 0.01 niter = 20 fcstep = 0 nstcgsteep = 1000 nbfgscorr = 10 nstxout = 0 nstvout = 0 nstfout = 0 nstcheckpoint = 1000 nstlog = 50 nstenergy = 50 nstxtcout = 50 xtc-precision = 1000 nstlist = 5 ns_type = grid pbc = xyz rlist = 1.00 domain-decomposition = no coulombtype = Shift rcoulomb-switch = 0 rcoulomb = 1.00 epsilon-r = 1 vdw-type = Shift rvdw-switch = 0 rvdw = 1.00 DispCorr = EnerPres table-extension = 1 fourierspacing = 0.12 fourier_nx = 0 fourier_ny = 0 fourier_nz = 0 pme_order = 4 ewald_rtol = 1e-05 ewald_geometry = 3d epsilon_surface = 0 optimize_fft = no gb_algorithm = Still nstgbradii = 1 rgbradii = 2 gb_saltconc = 0 implicit_solvent = No Tcoupl = berendsen tc-grps = System tau_t = 0.05 ref_t = 273 Pcoupl = no ;berendsen annealing = no gen_vel = yes gen_temp = 273 gen_seed = 1993 constraints = none ;all-bonds lincs-warnangle = 30 morse = no disre = No disre-weighting = Conservative disre-mixed = no disre-fc = 1000 disre-tau = 0 nstdisreout = 100 orire = no orire-fc = 0 orire-tau = 0 nstorireout = 100 dihre = No dihre-fc = 1000 dihre-tau = 0 nstdihreout = 100 free-energy = no init-lambda = 0 delta-lambda = 0 sc-alpha = 0 sc-sigma = 0.3 cos-acceleration = 0 ___ gmx-users mailing listgmx-users@gromacs.org http://www.gromacs.org/mailman/listinfo/gmx-users Please don't post (un)subscribe requests to the list. Use the www interface or send it to [EMAIL PROTECTED] Can't post? Read http://www.gromacs.org/mailing_lists/users.php
Re: [gmx-users] Simulating crystalls
I am trying to simulate a crystal too. I have a box size of .65 nm and there are 8 bcc crystals in the system. Lattice parameter = 0.287 nm. I get a segmentation fault after mdrun -v. Is an atom with co-ordinates 0.574 0.000 0.000 equivalent to 0.287 0.000 0.000 in such a system? What I am trying to ask is, does GROMACS consider these as 2 different atoms placed at the respective positions or do the 2 atoms overlap? What could be the reason for the segmentation fault? Is it because the atoms are too close to each other? - Karamyog.On 5/23/06, Mark Abraham [EMAIL PROTECTED] wrote: David van der Spoel wrote: Gale, Ella wrote: Thanks for the advice, but my force-field has no charges and hence no coulomb potential, so I want the atoms to interact with itself via the Van der Waals functions. I'm using 3.2.1 at the moment and there is no option to use any of the Ewald summation techniques. Is this something that has been added in the most recent version? Ewald variant are only for Coulomb so far. You can use a normal cutoff or a shifted cutoff.One or two of the early PME papers describe implementation details forLJ PME, but I am not immediately aware of a modern MD code that implements it. If you want it, shop around.Mark___gmx-users mailing listgmx-users@gromacs.org http://www.gromacs.org/mailman/listinfo/gmx-usersPlease don't post (un)subscribe requests to the list. Use thewww interface or send it to [EMAIL PROTECTED] .Can't post? Read http://www.gromacs.org/mailing_lists/users.php ___ gmx-users mailing listgmx-users@gromacs.org http://www.gromacs.org/mailman/listinfo/gmx-users Please don't post (un)subscribe requests to the list. Use the www interface or send it to [EMAIL PROTECTED] Can't post? Read http://www.gromacs.org/mailing_lists/users.php
Re: [gmx-users] Simulating crystalls
karamyog singh wrote: I am trying to simulate a crystal too. I have a box size of .65 nm and there are 8 bcc crystals in the system. Lattice parameter = 0.287 nm. I get a segmentation fault after mdrun -v. Is an atom with co-ordinates 0.574 0.000 0.000 equivalent to 0.287 0.000 0.000 in such a system? Cubic PBC with side length l means that (x,y,z) is equivalent to (x+al,y+bl,z+cl) for any integers a,b and c. What I am trying to ask is, does GROMACS consider these as 2 different atoms placed at the respective positions or do the 2 atoms overlap? No. GROMACS takes every atom you give in the .top file and uses it - there is no attempt to check for symmetry simplifications such as quantum chemistry codes do. What could be the reason for the segmentation fault? Is it because the atoms are too close to each other? You're on your own here unless you can provide more information. Mark ___ gmx-users mailing listgmx-users@gromacs.org http://www.gromacs.org/mailman/listinfo/gmx-users Please don't post (un)subscribe requests to the list. Use the www interface or send it to [EMAIL PROTECTED] Can't post? Read http://www.gromacs.org/mailing_lists/users.php
Re: [gmx-users] Simulating crystalls
Gale, Ella wrote: Hi, I was wondering is it possible to simulate a crystall by just simulating a unit cell in GROMACS? I want the molecule to be able to interact with the images of itself, but I don't seem to be able to set rlist, rvdw or rcoulomb any larger than half the unit cell size, which explicitly prevents my molecule from interacting with itself. Use Ewald, or preferably particle-mesh Ewald (PME) for crystalline systems. That was what they were invented for. Mark ___ gmx-users mailing listgmx-users@gromacs.org http://www.gromacs.org/mailman/listinfo/gmx-users Please don't post (un)subscribe requests to the list. Use the www interface or send it to [EMAIL PROTECTED] Can't post? Read http://www.gromacs.org/mailing_lists/users.php
RE: [gmx-users] Simulating crystalls
Title: RE: [gmx-users] Simulating crystalls Thanks for the advice, but my force-field has no charges and hence no coulomb potential, so I want the atoms to interact with itself via the Van der Waals functions. I'm using 3.2.1 at the moment and there is no option to use any of the Ewald summation techniques. Is this something that has been added in the most recent version? -Original Message- From: [EMAIL PROTECTED] on behalf of Mark Abraham Sent: Tue 5/23/2006 2:19 PM To: Discussion list for GROMACS users Subject: Re: [gmx-users] Simulating crystalls Gale, Ella wrote: Hi, I was wondering is it possible to simulate a crystall by just simulating a unit cell in GROMACS? I want the molecule to be able to interact with the images of itself, but I don't seem to be able to set rlist, rvdw or rcoulomb any larger than half the unit cell size, which explicitly prevents my molecule from interacting with itself. Use Ewald, or preferably particle-mesh Ewald (PME) for crystalline systems. That was what they were invented for. Mark ___ gmx-users mailing list gmx-users@gromacs.org http://www.gromacs.org/mailman/listinfo/gmx-users Please don't post (un)subscribe requests to the list. Use the www interface or send it to [EMAIL PROTECTED] Can't post? Read http://www.gromacs.org/mailing_lists/users.php ___ gmx-users mailing listgmx-users@gromacs.org http://www.gromacs.org/mailman/listinfo/gmx-users Please don't post (un)subscribe requests to the list. Use the www interface or send it to [EMAIL PROTECTED] Can't post? Read http://www.gromacs.org/mailing_lists/users.php
Re: [gmx-users] Simulating crystalls
Gale, Ella wrote: Thanks for the advice, but my force-field has no charges and hence no coulomb potential, so I want the atoms to interact with itself via the Van der Waals functions. I'm using 3.2.1 at the moment and there is no option to use any of the Ewald summation techniques. Is this something that has been added in the most recent version? Ewald variant are only for Coulomb so far. You can use a normal cutoff or a shifted cutoff. -Original Message- From: [EMAIL PROTECTED] on behalf of Mark Abraham Sent: Tue 5/23/2006 2:19 PM To: Discussion list for GROMACS users Subject: Re: [gmx-users] Simulating crystalls Gale, Ella wrote: Hi, I was wondering is it possible to simulate a crystall by just simulating a unit cell in GROMACS? I want the molecule to be able to interact with the images of itself, but I don't seem to be able to set rlist, rvdw or rcoulomb any larger than half the unit cell size, which explicitly prevents my molecule from interacting with itself. Use Ewald, or preferably particle-mesh Ewald (PME) for crystalline systems. That was what they were invented for. Mark ___ gmx-users mailing listgmx-users@gromacs.org http://www.gromacs.org/mailman/listinfo/gmx-users Please don't post (un)subscribe requests to the list. Use the www interface or send it to [EMAIL PROTECTED] Can't post? Read http://www.gromacs.org/mailing_lists/users.php ___ gmx-users mailing listgmx-users@gromacs.org http://www.gromacs.org/mailman/listinfo/gmx-users Please don't post (un)subscribe requests to the list. Use the www interface or send it to [EMAIL PROTECTED] Can't post? Read http://www.gromacs.org/mailing_lists/users.php -- David. David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics group, Dept. of Cell and Molecular Biology, Uppsala University. Husargatan 3, Box 596, 75124 Uppsala, Sweden phone: 46 18 471 4205 fax: 46 18 511 755 [EMAIL PROTECTED] [EMAIL PROTECTED] http://folding.bmc.uu.se ___ gmx-users mailing listgmx-users@gromacs.org http://www.gromacs.org/mailman/listinfo/gmx-users Please don't post (un)subscribe requests to the list. Use the www interface or send it to [EMAIL PROTECTED] Can't post? Read http://www.gromacs.org/mailing_lists/users.php
