{"id":1774,"date":"2020-02-17T09:12:41","date_gmt":"2020-02-17T00:12:41","guid":{"rendered":"https:\/\/www.ag.kagawa-u.ac.jp\/charlesy\/?p=1774"},"modified":"2024-02-02T15:56:29","modified_gmt":"2024-02-02T06:56:29","slug":"plumed-patched-gromacs%e3%81%ae%e3%82%a4%e3%83%b3%e3%82%b9%e3%83%88%e3%83%bc%e3%83%ab","status":"publish","type":"post","link":"https:\/\/www.ag.kagawa-u.ac.jp\/charlesy\/2020\/02\/17\/plumed-patched-gromacs%e3%81%ae%e3%82%a4%e3%83%b3%e3%82%b9%e3%83%88%e3%83%bc%e3%83%ab\/","title":{"rendered":"PLUMED-patched GROMACS\u306e\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3068\u30c6\u30b9\u30c8"},"content":{"rendered":"

\u7814\u7a76\u5ba4\u5185\u30e1\u30e2<\/p>\n

\u74b0\u5883: CentOS 7.
\nGROMACS 2019.4, PLUMED 2.6.0, GCC 6.3.1, OpenMPI 4.0.2, AmberTools 19.<\/p>\n

\u53c2\u8003\u30b5\u30a4\u30c8: Using Hamiltonian replica exchange with GROMACS – www.plumed.org<\/a><\/p>\n

\u53c2\u8003\u6587\u732e: Giovanni, B. Hamiltonian replica exchange in GROMACS: a flexible implementation.
\nMol. Phys.<\/span> 2014<\/span>, 112<\/span>, 379\u2013384. DOI: 10.1080\/00268976.2013.824126<\/a>
\nWang, L.; Friesner, R. A.; Berne, B. J. Replica Exchange with Solute Scaling: A More Efficient Version of Replica Exchange with Solute Tempering (REST2).
\nJ. Phys. Chem. B<\/span> 2011<\/span>, 115<\/span>, 9431-9438. DOI:10.1021\/jp204407d<\/a><\/p>\n

\u5fa9\u7fd2: \u6e29\u5ea6\u30ec\u30d7\u30ea\u30ab\u4ea4\u63dbMD<\/h2>\n

\"\"<\/p>\n

Temperature Replica Exchange Molecular Dynamics\uff08\u6e29\u5ea6\u30ec\u30d7\u30ea\u30ab\u4ea4\u63dbMD\u3001T-REMD\uff09\u6cd5\u3067\u306f\u3001\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u306elocal minimum\u9593\u306e\u969c\u58c1\u3092\u8d85\u3048\u3066\u52b9\u7387\u7684\u306b\u30b5\u30f3\u30d7\u30ea\u30f3\u30b0\u3092\u884c\u3046\u305f\u3081\uff08\u5c0f\u5206\u5b50\u3068\u30bf\u30f3\u30d1\u30af\u8cea\u306e\u7d50\u5408\u69d8\u5f0f\u3060\u3063\u305f\u308a\u3001\u30bf\u30f3\u30d1\u30af\u8cea\/\u30da\u30d7\u30c1\u30c9\u306e\u30b3\u30f3\u30db\u30e1\u30fc\u30b7\u30e7\u30f3\u3060\u3063\u305f\u308a\uff09\u3001\u6e29\u5ea6\u304c\u7570\u306a\u308b\u30ec\u30d7\u30ea\u30ab\u3092\u8907\u6570\u7528\u610f\u3057\u3066\u30e1\u30c8\u30ed\u30dd\u30ea\u30b9\uff1d\u30d8\u30a4\u30b9\u30c6\u30a3\u30f3\u30b0\u30ba\u57fa\u6e96\uff08Metropolis\u2013Hastings criterion\u3002\u78ba\u7387\u306f\u4e0b\u306e\u5f0f\u3067\u8a08\u7b97\u3055\u308c\u308b\uff09\u306b\u5f93\u3063\u3066\u30ec\u30d7\u30ea\u30ab\u306e\u5ea7\u6a19\u306e\u4ea4\u63db\u3092\u8a66\u884c\u3057\u307e\u3059\u3002<\/p>\n\\(\nP(1\u21942)=min\\left(1, \\exp \\left[\\left(\\frac{1}{k_BT_1}\u2212\\frac{1}{k_BT_2}\\right)(U_1\u2212U_2)\\right]\\right)
\n\\)\n

min\u306f\u5c0f\u3055\u3044\u65b9\u3092\u53d6\u308b\u95a2\u6570\u306a\u306e\u3067\u3001\u72b6\u614b1\u306e\u65b9\u304c\u6e29\u5ea6\u304c\u4f4e\u3044\u3068\u3057\u3066\u3001\u72b6\u614b1\u306e\u77ac\u9593\u7684\u306a\u30dd\u30c6\u30f3\u30b7\u30e3\u30eb\u30a8\u30cd\u30eb\u30ae\u30fc\uff08\\(U_1\\)\uff09\u304c\u72b6\u614b2\u306e\u30a8\u30cd\u30eb\u30ae\u30fc\uff08\\(U_2\\)\uff09\u4ee5\u4e0a\u306e\u5834\u5408\u306f\uff08\\(U_1 \\geq U_2\\)\uff09\u5fc5\u305a\u4ea4\u63db\u304c\u884c\u308f\u308c\uff08\u78ba\u73871= 100%\uff09\u3001\u72b6\u614b1\u306e\u65b9\u304c\u4f4e\u3051\u308c\u3070\u30a8\u30cd\u30eb\u30ae\u30fc\u5dee\u306b\u5fdc\u3058\u3066\u6307\u6570\u95a2\u6570\u7684\u306b\u5c0f\u3055\u304f\u306a\u308b\u78ba\u7387\u3067\u4ea4\u63db\u3055\u308c\u307e\u3059\u3002<\/p>\n

\u5727\u529b\u5236\u5fa1\u3092\u3057\u3066\u3044\u308b\u5834\u5408\u306e\u4ea4\u63db\u78ba\u7387\u306f\u3001<\/p>\n\\(\nP(1\u21942)=min\\left(1, \\exp \\left[\\left(\\frac{1}{k_BT_1}\u2212\\frac{1}{k_BT_2}\\right)(U_1\u2212U_2) + \\left(\\frac{P_1}{k_BT_1} – \\frac{P_2}{k_BT_2}\\right) (V_1 – V_2)\\right]\\right)
\n\\)\n

\u3068\u306a\u308a\u307e\u3059\u304c\u3001\u307b\u3068\u3093\u3069\u306e\u5834\u5408\u4f53\u7a4dV\u306e\u5dee\u304c\u5c0f\u3055\u3044\u306e\u3067\u7b2c2\u9805\u306f\u7121\u8996\u3067\u304d\u308b\u3002<\/p>\n

Gromacs\uff08MPI\u3092\u6709\u52b9\u3001”build_mdrun_only” \u30aa\u30d7\u30b7\u30e7\u30f3\u3092on\u306b\u3057\u3066make install\u3057\u305fmdrun_mpi\u3092\u4f7f\u7528\uff09\u3067\u306f\u3001\u4f8b\u3048\u30701\u304b\u30894\u306e\u6570\u5b57\u306e4\u3064\u306e\u30c7\u30a3\u30ec\u30af\u30c8\u30ea\u306b\u305d\u308c\u305e\u308c\u30d5\u30a1\u30a4\u30eb\u540dtestrun.tpr\u306e\u5b9f\u884c\u30d5\u30a1\u30a4\u30eb\u304c\u7528\u610f\u3055\u308c\u3066\u3044\u308b\u3068\u3059\u308b\u3068<\/p>\n

$ mpirun -np 4 --host localhost:4 mdrun_mpi -deffnm testrun -multidir [1234] -replex 500<\/code><\/pre>\n
\u5168\u3066\u306e\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u306e\u6e29\u5ea6\u304c\u540c\u3058\u3067\u3042\u308c\u3070\u3001100%\u8fd1\u304f\u4ea4\u63db\u304c\u8d77\u3053\u308b\u306f\u305a\u3067\u3059\u3002\u4e00\u822c\u306b\u4ea4\u63db\u53d7\u5bb9\u78ba\u7387\u306f0.2\u7a0b\u5ea6\u304c\u671b\u307e\u3057\u3044\u305d\u3046\u3067\u3059\u3002\u5404\u8a2d\u5b9a\u6e29\u5ea6\u306fTemperature generator for REMD-simulations<\/a>\u3067\u8a08\u7b97\u3057\u3066\u304f\u308c\u307e\u3059\u3002\u305f\u3060\u3057\u3053\u306e\u30b5\u30a4\u30c8\u306b\u5165\u529b\u3057\u305f\u78ba\u7387\u306b\u306f\u5b9f\u969b\u306b\u306f\u5fc5\u305a\u3057\u3082\u306a\u3089\u306a\u3044\u306e\u3067\u3001\u77ed\u3044\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u3092\u8d70\u3089\u305b\u3066\u4ea4\u63db\u78ba\u7387\u3092\u78ba\u8a8d\u3057\u3001\u4f55\u5ea6\u304b\u8abf\u6574\u3092\u884c\u3046\u5fc5\u8981\u304c\u3042\u308b\u3067\u3057\u3087\u3046\u3002<\/p>\n
1\u3064\u306e\u30ec\u30d7\u30ea\u30ab\u306b\u7740\u76ee\u3059\u308b\u3068\u3001\u7570\u306a\u308b\u6e29\u5ea6\u3092\u30e9\u30f3\u30c0\u30e0\u30a6\u30a9\u30fc\u30af\uff08\u9154\u6b69\uff09\u3057\u3066\u305d\u308c\u306b\u3088\u3063\u3066\u30a8\u30cd\u30eb\u30ae\u30fc\u969c\u58c1\u3092\u8d85\u3048\u308b\u306e\u3067\u3001Simulated Annealing\uff08\u64ec\u4f3c\u713c\u304d\u306a\u307e\u3057\uff09\u6cd5\u3068\u985e\u4f3c\u3057\u3066\u3044\u307e\u3059\uff08\u30ec\u30d7\u30ea\u30ab\u4ea4\u63db\u306e\u5834\u5408\u306f\u30a2\u30cb\u30ea\u30fc\u30f3\u30b0\u3067\u306f\u306a\u304f\u30c6\u30f3\u30d1\u30ea\u30f3\u30b0\u3008tempering\u3001\u713c\u304d\u3082\u3069\u3057\u3009\u3068\u547c\u3070\u308c\u308b\uff09\u3002\u8907\u6570\u306e\u30c6\u30f3\u30d1\u30ea\u30f3\u30b0\u3092\u540c\u6642\u306b\u884c\u3046\u305f\u3081\u3001T-REMD\u306fParallel Tempering\uff08\u5e73\u884c\u713c\u304d\u3082\u3069\u3057\uff09\u3068\u547c\u3070\u308c\u307e\u3059\u3002Simulated Annealing\u6cd5\u3068REMD\u6cd5\u3067\u306fREMD\u306e\u65b9\u304c\u8a08\u7b97\u30b3\u30b9\u30c8\u304c\u304b\u306a\u308a\u9ad8\u304f\u306a\u308a\u307e\u3059\u304c\u3001T-REMD\u306e\u5229\u70b9\u306f\u3001\u305d\u308c\u305e\u308c\u306e\u6e29\u5ea6\u3067\u30a2\u30f3\u30b5\u30f3\u30d6\u30eb\uff08\u7d71\u8a08\u96c6\u56e3\uff09\u306b\u4e00\u81f4\u3059\u308b\u30c8\u30e9\u30b8\u30a7\u30af\u30c8\u30ea\u3092\u751f\u6210\u3059\u308b\u70b9\u3067\u3059\uff08\u539f\u5b50\u306e\u5ea7\u6a19\u3092\u4ea4\u63db\u3059\u308b\u306e\u3067\u69cb\u9020\u306f\u9014\u4e2d\u3067\u98db\u3093\u3067\u3044\u308b\uff09\u3002<\/p>\n
\u30cf\u30df\u30eb\u30c8\u30cb\u30a2\u30f3\u30ec\u30d7\u30ea\u30ab\u4ea4\u63db\u6cd5<\/h2>\n
\u89e3\u8aac\u30fbPLUMED\u306e\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb<\/h3>\n
Hamiltonian Replica Exchange (HREX) \u6cd5\u306f\u6e29\u5ea6\u3067\u306f\u306a\u304f\u529b\u5834\u306e\u30d1\u30e9\u30e1\u30fc\u30bf\u304c\u7570\u306a\u308b\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u3092\u8907\u6570\u4e26\u5217\u5b9f\u884c\u3057\u3066\u3001\u305d\u308c\u3089\u306e\u9593\u3067\u4ea4\u63db\u3092\u884c\u3046\u65b9\u6cd5\u3067\u3059\u3002\u6eb6\u8cea\u3068\u6c34\u306e\u30d1\u30e9\u30e1\u30fc\u30bf\u3092\u6bb5\u968e\u7684\u306b\u5f31\u3081\u305f\u30ec\u30d7\u30ea\u30ab\u3092\u4f7f\u3046Replica Exchange with Solute Scaling\uff08REST2\uff09\u306f\u3001T-REMD\u306b\u6bd4\u3079\u3066\u5c11\u306a\u3044\u30ec\u30d7\u30ea\u30ab\u6570\u3067\u52b9\u7387\u7684\u306b\u30b3\u30f3\u30db\u30e1\u30fc\u30b7\u30e7\u30f3\u306e\u30b5\u30f3\u30d7\u30ea\u30f3\u30b0\u304c\u3067\u304d\u308b\u3053\u3068\u304c\u793a\u3055\u308c\u3066\u3044\u307e\u3059\u3002
\n\u203b\u6c34\u306e\u6e29\u5ea6\u306f\u51b7\u305f\u3044\u307e\u307e\u3067\u3001\u6eb6\u8cea\u306e\u6e29\u5ea6\u3060\u3051\u4e0a\u3052\u305fReplica Exchange with Solute Tempering\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u304c “REST” \u3068\u547c\u3070\u308c\u308b\u306e\u306b\u5bfe\u3057\u3066\u3001”Hamiltonian” Replica Exchange with Solute Scaling\u306f “REST2” \u3068\u547c\u3070\u308c\u308b\u3002<\/p>\n
Gromacs\u306b\u306f2\u3064\u306e\u72b6\u614b\u306e\u81ea\u7531\u30a8\u30cd\u30eb\u30ae\u30fc\u5dee\u3092\u8a08\u7b97\u3059\u308b\u305f\u3081\u306eBAR\uff08Bennett Acceptance Ratio\u3001\u30d9\u30cd\u30c3\u30c8\u53d7\u5bb9\u6bd4\uff09\u6cd5\u304c\u5b9f\u88c5\u3055\u308c\u3066\u3044\u3066\u3001\u305d\u306e\u6642\u306f\u6eb6\u8cea\uff08\u5c0f\u5206\u5b50\u3084\u30bf\u30f3\u30d1\u30af\u8cea\uff09\u3068\u7cfb\u3068\u306e\u30af\u30fc\u30ed\u30f3\u76f8\u4e92\u4f5c\u7528\u3084\u30d5\u30a1\u30f3\u30c7\u30eb\u30ef\u30fc\u30eb\u30b9\u76f8\u4e92\u4f5c\u7528\u3092\u6bb5\u968e\u7684\u306b\u6d88\u53bb\u3057\u3066\u3044\u304d\u307e\u3059\u3002\u3053\u308c\u3092\u4f7f\u3063\u3066\u3001\u4f8b\u3048\u3070\u30ea\u30ac\u30f3\u30c9\u3068\u5468\u56f2\u3068\u306e\u76f8\u4e92\u4f5c\u7528\u3092\u5f31\u3081\u305f\u30ec\u30d7\u30ea\u30ab\u3092\u4f5c\u308b\u3053\u3068\u3067\u3001\u30a8\u30cd\u30eb\u30ae\u30fc\u969c\u58c1\u3092\u8d85\u3048\u305f\u72b6\u614b\u9593\u306e\u9077\u79fb\u3092\u5bb9\u6613\u306b\u3057\u307e\u3059\u3002\u305f\u3060\u3057\u3068\u3066\u3082\u9045\u3044\uff08\u5f8c\u8ff0\uff09\u3002<\/p>\n
\u5225\u306e\u65b9\u6cd5\u3068\u3057\u3066\u3001PLUMED\u30d7\u30ed\u30b0\u30e9\u30e0\u3068PLUMED\u3067\u30d1\u30c3\u30c1\u3092\u5f53\u3066\u305fGROMACS\u3092\u4f7f\u7528\u3059\u308b\u65b9\u6cd5\u304c\u3042\u308a\u307e\u3059\u3002\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u6cd5\u306a\u3069\u306fInstallation – www.plumed.org<\/a>\u306b\u8a73\u3057\u304f\u89e3\u8aac\u3057\u3066\u3042\u308a\u307e\u3059\u3002<\/p>\n
$ source \/usr\/local\/gromacs\/2019.4_plumed\/bin\/GMXRC\n<\/code><\/pre>\n
\u30d1\u30c3\u30c1\u3092\u5f53\u3066\u3066\u30d3\u30eb\u30c9\u3057\u305fGROMACS\u306fHREX\u6cd5\u306e\u305f\u3081\u306e “mdrun -hrex” \u30aa\u30d7\u30b7\u30e7\u30f3\u304c\u4f7f\u3048\u308b\u3088\u3046\u306b\u306a\u3063\u3066\u3044\u308b\u306f\u305a\u3067\u3059\uff08gmx_mpi mdrun -h\u3067\u78ba\u8a8d\uff09\u3002<\/p>\n
\u6e96\u5099<\/h3>\n
\u69cb\u9020\u30d5\u30a1\u30a4\u30eb\u306e\u4f5c\u6210\u3068\u5e73\u8861\u5316<\/h4>\n
AMBER Tutorial A19<\/a>\u3092\u53c2\u8003\u306bAmber\u3092\u4f7f\u3063\u3066alanine\u3092\u4f5c\u308b\u3002<\/p>\n
$ source \/usr\/local\/amber18\/\/amber.sh\n$ tleap\n> source leaprc.protein.ff03ua\n> m = sequence { ACE ALA NME }\n> savepdb m diala.pdb\n> quit\n$ gmx_mpi pdb2gmx -f diala.pdb -o conf.gro -ignh\n> 1 [enter]\n> 1 [enter]\n$ gmx_mpi editconf -f conf.gro -o box.gro -bt cubic -d 1\n$ gmx_mpi solvate -cp box.gro -p topol.top -o solv.gro\n<\/code><\/pre>\n
\u30a2\u30df\u30ce\u9178\u306e\u539f\u5b50\u657022\u3001\u6c34\u5206\u5b50\u6570752\u3002<\/p>\n
\u5fc5\u8981\u306a\u30d5\u30a1\u30a4\u30eb\u306e\u4e2d\u8eab:
\nminim.mdp<\/p>\n
; minim.mdp - used as input into grompp to generate em.tpr\n; Parameters describing what to do, when to stop and what to save\nintegrator  = steep         ; Algorithm (steep = steepest descent minimization)\nemtol       = 1000.0        ; Stop minimization when the maximum force < 1000.0 kJ/mol/nm\nemstep      = 0.01          ; Minimization step size\nnsteps      = 50000         ; Maximum number of (minimization) steps to perform\n\n; Parameters describing how to find the neighbors of each atom and how to calculate the interactions\nnstlist         = 1         ; Frequency to update the neighbor list and long range forces\ncutoff-scheme   = Verlet    ; Buffered neighbor searching\nns_type         = grid      ; Method to determine neighbor list (simple, grid)\ncoulombtype     = PME       ; Treatment of long range electrostatic interactions\nrcoulomb        = 1.0       ; Short-range electrostatic cut-off\nrvdw            = 1.0       ; Short-range Van der Waals cut-off\npbc             = xyz       ; Periodic Boundary Conditions in all 3 dimensions<\/pre>\n
nvt.mdp<\/p>\n
define      = -DPOSRES; position restrain the protein\n; Run parameters\nintegrator  = md        ; leap-frog integrator\nnsteps      = 50000     ; 2 * 50000 = 100 ps\ndt          = 0.002     ; 2 fs\n; Output control\nnstxout     = 5000       ; save coordinates every 1.0 ps\nnstvout     = 5000       ; save velocities every 1.0 ps\nnstenergy   = 5000       ; save energies every 1.0 ps\nnstlog      = 5000       ; update log file every 1.0 ps\nnstxtcout     = 5000 \n;energygrps  = Protein \n; Bond parameters\ncontinuation    = no           ; first dynamics run\nconstraint_algorithm = lincs    ; holonomic constraints \nconstraints     = h-bonds     ; all bonds (even heavy atom-H bonds) constrained\nlincs_iter      = 1             ; accuracy of LINCS\nlincs_order     = 4             ; also related to accuracy\n; Neighborsearching\ncutoff-scheme   = Verlet\n;vdwtype =cutoff\n;vdw-modifier = force-switch\nrlist =1\nrcoulomb        = 1       ; short-range electrostatic cutoff (in nm)\nrvdw            = 1       ; short-range van der Waals cutoff (in nm)\n;rvdw-switch =1.0\n; Electrostatics\ncoulombtype     = PME       ; Particle Mesh Ewald for long-range electrostatics\npme_order       = 4         ; cubic interpolation\nfourierspacing  = 0.16      ; grid spacing for FFT\n; Temperature coupling\ntcoupl      = V-rescale                     ; modified Berendsen thermostat\ntc-grps     = Protein Water    ; two coupling groups - more accurate\ntau_t       = 0.1   0.1                     ; time constant, in ps\nref_t       = 310   310                    ; reference temperature, one for each group, in K\n;refcoord_scaling    = com\n; Periodic boundary conditions\npbc         = xyz       ; 3-D PBC\n; Dispersion correction\nDispCorr    = EnerPres  ; account for cut-off vdW scheme\n; Velocity generation\ngen_vel     = yes        ; velocity generation off after NVT<\/pre>\n
npt.mdp<\/p>\n
define                  = -DPOSRES  ; position restrain the protein\n; Run parameters\nintegrator              = md        ; leap-frog integrator\nnsteps                  = 50000     ; 2 * 50000 = 100 ps\ndt                      = 0.002     ; 2 fs\n; Output control\nnstxout                 = 500       ; save coordinates every 1.0 ps\nnstvout                 = 500       ; save velocities every 1.0 ps\nnstenergy               = 500       ; save energies every 1.0 ps\nnstlog                  = 500       ; update log file every 1.0 ps\n; Bond parameters\ncontinuation            = yes       ; Restarting after NVT \nconstraint_algorithm    = lincs     ; holonomic constraints \nconstraints             = h-bonds   ; bonds involving H are constrained\nlincs_iter              = 1         ; accuracy of LINCS\nlincs_order             = 4         ; also related to accuracy\n; Nonbonded settings \ncutoff-scheme           = Verlet    ; Buffered neighbor searching\nns_type                 = grid      ; search neighboring grid cells\nnstlist                 = 10        ; 20 fs, largely irrelevant with Verlet scheme\nrcoulomb                = 1.0       ; short-range electrostatic cutoff (in nm)\nrvdw                    = 1.0       ; short-range van der Waals cutoff (in nm)\nDispCorr                = EnerPres  ; account for cut-off vdW scheme\n; Electrostatics\ncoulombtype             = PME       ; Particle Mesh Ewald for long-range electrostatics\npme_order               = 4         ; cubic interpolation\nfourierspacing          = 0.16      ; grid spacing for FFT\n; Temperature coupling is on\ntcoupl                  = V-rescale             ; modified Berendsen thermostat\ntc-grps                 = Protein Water   ; two coupling groups - more accurate\ntau_t                   = 0.1     0.1           ; time constant, in ps\nref_t                   = 310     310           ; reference temperature, one for each group, in K\n; Pressure coupling is on\npcoupl                  = Parrinello-Rahman     ; Pressure coupling on in NPT\npcoupltype              = isotropic             ; uniform scaling of box vectors\ntau_p                   = 2.0                   ; time constant, in ps\nref_p                   = 1.0                   ; reference pressure, in bar\ncompressibility         = 4.5e-5                ; isothermal compressibility of water, bar^-1\nrefcoord_scaling        = com\n; Periodic boundary conditions\npbc                     = xyz       ; 3-D PBC\n; Velocity generation\ngen_vel                 = no        ; Velocity generation is off<\/pre>\n
hrex.mdp<\/p>\n
define                  = \n; Run parameters\nintegrator              = md        ; leap-frog integrator\nnsteps                  = 500000     ; 2 * 500000 = 1000 ps\ndt                      = 0.002     ; 2 fs\n; Output control\nnstxout                 = 500       ; save coordinates every 1.0 ps\nnstvout                 = 500       ; save velocities every 1.0 ps\nnstenergy               = 500       ; save energies every 1.0 ps\nnstlog                  = 500       ; update log file every 1.0 ps\n; Bond parameters\ncontinuation            = yes       ; Restarting after NVT \nconstraint_algorithm    = lincs     ; holonomic constraints \nconstraints             = h-bonds   ; bonds involving H are constrained\nlincs_iter              = 1         ; accuracy of LINCS\nlincs_order             = 4         ; also related to accuracy\n; Nonbonded settings \ncutoff-scheme           = Verlet    ; Buffered neighbor searching\nns_type                 = grid      ; search neighboring grid cells\nnstlist                 = 10        ; 20 fs, largely irrelevant with Verlet scheme\nrcoulomb                = 1.0       ; short-range electrostatic cutoff (in nm)\nrvdw                    = 1.0       ; short-range van der Waals cutoff (in nm)\nDispCorr                = EnerPres  ; account for cut-off vdW scheme\n; Electrostatics\ncoulombtype             = PME       ; Particle Mesh Ewald for long-range electrostatics\npme_order               = 4         ; cubic interpolation\nfourierspacing          = 0.16      ; grid spacing for FFT\n; Temperature coupling is on\ntcoupl                  = V-rescale             ; modified Berendsen thermostat\ntc-grps                 = Protein Water   ; two coupling groups - more accurate\ntau_t                   = 0.1     0.1           ; time constant, in ps\nref_t                   = 310     310           ; reference temperature, one for each group, in K\n; Pressure coupling is on\npcoupl                  = Parrinello-Rahman     ; Pressure coupling on in NPT\npcoupltype              = isotropic             ; uniform scaling of box vectors\ntau_p                   = 2.0                   ; time constant, in ps\nref_p                   = 1.0                   ; reference pressure, in bar\ncompressibility         = 4.5e-5                ; isothermal compressibility of water, bar^-1\nrefcoord_scaling        = com\n; Periodic boundary conditions\npbc                     = xyz       ; 3-D PBC\n; Velocity generation\ngen_vel                 = no        ; Velocity generation is off<\/pre>\n
equilibration.sh<\/p>\n
#!/bin/bash\nsource /usr/local/gromacs/2019.4_plumed/bin/GMXRC\ngmx_mpi grompp -f minim.mdp -c solv.gro -o minim\ngmx_mpi mdrun -deffnm minim\ngmx_mpi grompp -f nvt.mdp -c minim.gro -r minim.gro -o nvt\ngmx_mpi mdrun -deffnm nvt\ngmx_mpi grompp -f npt.mdp -c nvt.gro -r nvt.gro -t nvt.cpt -o npt\ngmx_mpi mdrun -deffnm npt<\/pre>\n
$ sh .\/equilibration.sh<\/code><\/pre>\n
HREX\u7528\u30c8\u30dd\u30ed\u30b8\u30fc\u306e\u4f5c\u6210<\/h4>\n
Gromacs\u306e\u30c8\u30dd\u30ed\u30b8\u30fc\u30d5\u30a1\u30a4\u30eb\u306f\u5927\u62b5 “#include ~~.itp” \u306a\u3069\u3068\u5916\u90e8\u30d5\u30a1\u30a4\u30eb\u3092\u8aad\u307f\u8fbc\u3093\u3044\u308b\u7b87\u6240\u304c\u3042\u308a\u307e\u3059\u304c\u3001plumed\u3067\u51e6\u7406\u3059\u308b\u305f\u3081\u306b\u306f\u5168\u3066\u306e\u30d1\u30e9\u30e1\u30fc\u30bf\u304c\u66f8\u304d\u8fbc\u307e\u308c\u305f\u30c8\u30dd\u30ed\u30b8\u30fc\u30d5\u30a1\u30a4\u30eb\u304c\u5fc5\u8981\u3067\u3059\u3002\u3053\u308c\u306f\u3001grompp -pp \u30aa\u30d7\u30b7\u30e7\u30f3\u3067\u751f\u6210\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3059\uff08processed.top\u30d5\u30a1\u30a4\u30eb\u304c\u51fa\u529b\u3055\u308c\u308b\uff09\u3002<\/p>\n
$ gmx_mpi grompp -f minim.mdp -c npt.gro -pp\n<\/code><\/pre>\n
\u51fa\u529b\u3055\u308c\u305fprocessed.top\u306e\u4e2d\u3067\u529b\u5834\u30d1\u30e9\u30e1\u30fc\u30bf\u3092\u30b9\u30b1\u30fc\u30ea\u30f3\u30b0\u3057\u305f\u3044\u5206\u5b50\uff08\u6c34\u4e2d\u306e\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u306a\u3089\u5927\u62b5\u306f\u6eb6\u8cea\uff09\u306eatom type\u306e\u5f8c\u308d\u306b\u30a2\u30f3\u30c0\u30fc\u30d0\u30fc\u3092\u4ed8\u52a0\u3057\u307e\u3059\uff08\u53c2\u8003: https:\/\/www.plumed.org\/doc-v2.5\/user-doc\/html\/hrex.html<\/a>\uff09\u3002<\/p>\n
[ moleculetype ]\n; Name            nrexcl\nProtein             3\n\n[ atoms ]\n;   nr       type  resnr residue  atom   cgnr     charge       mass  typeB    chargeB      massB\n; residue   1 ACE rtp ACE  q  0.0\n     1         CT_      1    ACE    CH3      1  -0.190264      12.01\n     2         HC_      1    ACE   HH31      2    0.07601      1.008\n     3         HC_      1    ACE   HH32      3   0.076011      1.008\n     4         HC_      1    ACE   HH33      4    0.07601      1.008\n     5          C_      1    ACE      C      5   0.512403      12.01<\/pre>\n
Position restraint\u306e\u60c5\u5831\u3082\u5168\u3066\u30a4\u30f3\u30af\u30eb\u30fc\u30c9\u3055\u308c\u3066\u3044\u308b\u306e\u3067\u524a\u9664\u3057\u307e\u3059\u3002<\/p>\n
\u6b21\u306b\u3001\u4ee5\u4e0b\u306escript (scaling.py) \u3092\u5b9f\u884c\u3057\u307e\u3059\u3002Python wrapper\u304c\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u6e08\u307f\u306e\u5834\u5408\u306f”import plumed”\u3067\u4f7f\u3048\u307e\u3059\u3002<\/p>\n
#! /usr/bin/env python3\nimport math, os, glob, subprocess\n\n# \u30ec\u30d7\u30ea\u30ab\u306e\u6570\nnrep = 4\n# "\u6709\u52b9" \u6e29\u5ea6\u7bc4\u56f2\ntmin = 310\ntmax = 1200\n\n# build  geometric progression\ntemp_list = [ tmin * math.exp(i * math.log( tmax / tmin ) / ( nrep - 1 )) for i in range( nrep )]\nlambda_list = [ temp_list[0] / temp_list[i] for i in range( nrep )]\nprint("temperature :", temp_list)\nprint("lambda :", lambda_list)\n\n# \u4e0d\u8981\u306a\u30d5\u30a1\u30a4\u30eb\u306e\u524a\u9664\nfiles = glob.glob( "./#*" )\nfiles += glob.glob( "./topol*" )\n\nfor f in files:\n    os.remove( f )\n\nfor i in range( nrep ):\n    command1 = ["mkdir", "replica"+str(i)]\n    subprocess.call( command1 )\n    # lambda\u306e\u5024\u3092T[0]/T[i] \u3068\u3057\u3066\u9078\u3076\u3002\n    # \u9ad8\u3044\u6e29\u5ea6\u306f\u4f4e\u3044lambda\u5024\u306b\u76f8\u5f53\u3059\u308b\u3002\n    lambdavalue = temp_list[0] / temp_list[i]\n    # \u30c8\u30dd\u30ed\u30b8\u30fc\u306e\u51e6\u7406\n    # \u7d50\u679c\u306f "diff topol0.top topol1.top" \u3067\u78ba\u8a8d\u3067\u304d\u308b\u3002\n    command2 = [ "plumed", "partial_tempering", str( lambdavalue ), "<", "processed.top", ">", "topol"+str(i)+".top" ]\n    subprocess.call( command2 )\n    command3 = [ "mv",  "topol"+str(i)+".top", "./replica"+str( i )]\n    subprocess.call( command3 )\n\n    # tpr\u30d5\u30a1\u30a4\u30eb\u306e\u6e96\u5099\n    # \u6eb6\u8cea\u304c\u96fb\u8377\u3092\u6301\u3063\u3066\u3044\u305f\u5834\u5408\u3001\u90e8\u5206\u96fb\u8377\u3092\u30b9\u30b1\u30fc\u30ea\u30f3\u30b0\u3057\u3066\u3044\u304f\u3068\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u7cfb\u306e\u96fb\u8377\u304c0\u3067\u306f\u306a\u304f\u306a\u308a\u8b66\u544a\u304c\u51fa\u308b\u306e\u3067 -maxwarn 1 \u3067\u56de\u907f\u3059\u308b\u3002\n    command4 = [ "gmx_mpi", "grompp", "-maxwarn", "1", "-o", "replica"+str( i )+"/hrex.tpr", "-f", "hrex.mdp", "-c", "npt.gro", "-p", "replica"+str( i )+"/topol"+str(i)+".top" ]\n    subprocess.call( command4 )\n    command5 = [ "touch",  "replica"+str( i )+"/hrex.dat"] #\u7a7a\u30d5\u30a1\u30a4\u30eb\u306e\u4f5c\u6210\n    subprocess.call( command5 )\n<\/pre>\n
$ python3 .\/scaling.py<\/code><\/pre>\n
REST2\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u306e\u5b9f\u884c\uff08PLUMED\u30011 ns\uff09<\/h3>\n
$ mpirun -np 4  gmx_mpi mdrun -deffnm hrex -ntomp 1 -multidir replica{0..3} -replex 500 -hrex -plumed hrex.dat -pin on \n<\/code><\/pre>\n
Lambda\u306e\u5024: [1.0, 0.532, 0.353, 0.258]\u3002<\/p>\n
\u30ed\u30b0\u30d5\u30a1\u30a4\u30eb\u3092\u898b\u308b\u3068\uff08replica*\/hrex.log\uff09\u3001\u30ec\u30d7\u30ea\u30ab\u306e\u4ea4\u63db\u53d7\u5bb9\u78ba\u7387\u304c\u5747\u4e00\u3067\u306f\u306a\u304f\u3001\u300c\u9ad8\u6e29\u300d\u5074\u306b\u3044\u304f\u307b\u3069\u9ad8\u304f\u306a\u3063\u3066\u3057\u307e\u3063\u3066\u3044\u305f\u3002<\/p>\n
Repl  average number of exchanges:\nRepl     0    1    2    3\nRepl      .13  .21  .33<\/pre>\n
\u30b9\u30b1\u30fc\u30ea\u30f3\u30b0\u3092
\nlambda : [1.0, 0.706, 0.446, 0.258]\n\u306b\u8abf\u6574\u3057\u305f\u3068\u3053\u308d\u3001<\/p>\n
Repl  average number of exchanges:\nRepl     0    1    2    3\nRepl      .26  .21  .21<\/pre>\n
\u3068\u306a\u3063\u305f\u3002\u4ea4\u63db\u53d7\u5bb9\u78ba\u7387\u306f20%\u7a0b\u5ea6\u304c\u9069\u5207\u3068\u306e\u3053\u3068\uff08\u53c2\u8003: \u6e29\u5ea6\u30ec\u30d7\u30ea\u30ab\u4ea4\u63db\u5206\u5b50\u52d5\u529b\u5b66\u6cd5\u306b\u3064\u3044\u3066\u306e Q&A\uff08PDF\u30d5\u30a1\u30a4\u30eb\uff09<\/a>\uff09\u306a\u306e\u3067\u6982\u306d\u826f\u597d\u3068\u8a00\u3048\u308b\u3002<\/p>\n
Performance\u306f120.127 ns\/day\u3060\u3063\u305f\u3002<\/p>\n
GROMACS\u5358\u72ec\u306e\u6a5f\u80fd\u3092\u4f7f\u3063\u305fREST2\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u306e\u5b9f\u884c\uff08100 ps\uff09<\/h3>\n
\u53c2\u8003: Free energy of solvation tutorial – gromacs.org<\/a><\/p>\n
run.mdp<\/p>\n
; using sd integrator with 50000 time steps (100 ps)\nintegrator = sd ; stochastic (velocity Langevin) dynamics integrator\u3002\u30e9\u30f3\u30b8\u30e5\u30d0\u30f3\u52d5\u529b\u5b66\u3067\u306f\u6e29\u5ea6\u5236\u5fa1\u3092\u884c\u3048\u308b\u306e\u3067\u3001tcoupl\u30d1\u30e9\u30e1\u30fc\u30bf\u306f\u7121\u8996\u3055\u308c\u308b\u3002\nnsteps = 50000\ndt = 0.002\nnstenergy = 1000\nnstlog = 5000\n; cut-offs at 1.0nm\ncutoff-scheme = Verlet\nrlist = 1.0\ndispcorr = EnerPres\nvdw-type = pme\nrvdw = 1.0\n; Coulomb interactions\ncoulombtype = pme\nrcoulomb = 1.0\nfourierspacing = 0.12\n; Constraints\nconstraints = all-bonds\n; set temperature to 310K\n; tcoupl = v-rescale ; \u5fc5\u8981\u306a\u3044\ntc-grps = system\ntau-t = 0.2\nref-t = 310\n; set pressure to 1 bar with a thermostat that gives a correct\n; thermodynamic ensemble\npcoupl = parrinello-rahman \nref-p = 1\ncompressibility = 4.5e-5\ntau-p = 5\n; and set the free energy parameters\nfree-energy = yes\ncouple-moltype = Protein\n; these 'soft-core' parameters make sure we never get overlapping\n; charges as lambda goes to 0\nsc-power = 1\nsc-sigma = 0.3\nsc-alpha = 1.0\n; we still want the molecule to interact with itself at lambda=0\ncouple-intramol = no\ncouple-lambda1 = vdwq ; VdW\u529b\u3068\u30af\u30fc\u30ed\u30f3\u529b\u306e\u4e21\u65b9\u3092\u30b9\u30b1\u30fc\u30ea\u30f3\u30b0\u3059\u308b\u3002\ncouple-lambda0 = none\ninit-lambda-state = $LAMBDA$ \n; These are the lambda states at which we simulate\n; for separate LJ and Coulomb decoupling, use\nfep-lambdas = 0.258 0.446 0.706 1.0 ; \u524d\u7bc0\u3068\u540c\u3058\u03bb\u5024<\/pre>\n
$ (for i in {0..3}; do mkdir lambda_$i; cp run.mdp topol.top npt.gro lambda_$i\/; sed -i -e "s\/\\\\\\$LAMBDA\\\\\\$\/$i\/" lambda_$i\/run.mdp; done)\n$ mpirun -np 4 gmx_mpi mdrun -deffnm fep -multidir lambda_{0..3} -replex 500 -pin on -ntomp 1 -v<\/code><\/pre>\n
\u7d50\u679c:<\/p>\n
Repl  average number of exchanges:\nRepl     0    1    2    3\nRepl      .16  .00  .00<\/pre>\n
\u3068\u3046\u307e\u304f\u4ea4\u63db\u3067\u304d\u306a\u304b\u3063\u305f\u3002fep\u6a5f\u80fd\u3092\u4f7f\u3063\u305f\u5834\u5408\u3001\u03bb\u306e\u5dee\u3092\u3082\u3063\u3068\u5c0f\u3055\u304f\u3059\u308b\u5fc5\u8981\u304c\u3042\u308b\u3002<\/p>\n
\u8a08\u7b97\u306f “Performance: 27.874 ns\/day” \u3068PLUMED\u306e4.3\u500d\u9045\u304b\u3063\u305f\u3002PLUMED\u3092\u4f7f\u3063\u305fREST2\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u3092integrator = sd\u3067\u5b9f\u884c\u3057\u305f\u3089\u3001integrator = md\u306b\u6bd4\u3079\u3066\u5c11\u3057\u9045\u3044\u3050\u3089\u3044\uff08111.273 ns\/day\uff09\u3060\u3063\u305f\u306e\u3067\u305d\u306e\u5f71\u97ff\u3067\u306f\u306a\u3044\u3002<\/p>\n
lambda\u304c0\u306b\u306a\u308b\u3053\u3068\u306f\u306a\u3044\u306e\u3067 “soft-core” \u30d1\u30e9\u30e1\u30fc\u30bf\u30923\u3064\u3068\u3082\u30b3\u30e1\u30f3\u30c8\u30a2\u30a6\u30c8\u3057\u3066\u5b9f\u884c\u3057\u305f\u3068\u3053\u308d\u3001\u30b9\u30d4\u30fc\u30c9\u306f30.148 ns\/day\u3001\u4ea4\u63db\u78ba\u7387\u306f<\/p>\n
Repl  average number of exchanges:\nRepl     0    1    2    3\nRepl      .16  .12  .10<\/pre>\n
\u3060\u3063\u305f\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"
\u7814\u7a76\u5ba4\u5185\u30e1\u30e2 \u74b0\u5883: CentOS 7. GROMACS 2 …<\/p>\n","protected":false},"author":1,"featured_media":2213,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"om_disable_all_campaigns":false,"_uag_custom_page_level_css":"","_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"_uf_show_specific_survey":0,"_uf_disable_surveys":false,"_locale":"ja","_original_post":"https:\/\/www.ag.kagawa-u.ac.jp\/charlesy\/?p=1774","footnotes":""},"categories":[5],"tags":[],"class_list":["post-1774","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-5","ja"],"aioseo_notices":[],"uagb_featured_image_src":{"full":["https:\/\/www.ag.kagawa-u.ac.jp\/charlesy\/wp\/wp-content\/uploads\/2020\/02\/Schematic_of_a_replica_exchange_molecular_dynamics_simulation.svg_.png",640,273,false],"thumbnail":["https:\/\/www.ag.kagawa-u.ac.jp\/charlesy\/wp\/wp-content\/uploads\/2020\/02\/Schematic_of_a_replica_exchange_molecular_dynamics_simulation.svg_-150x150.png",150,150,true],"medium":["https:\/\/www.ag.kagawa-u.ac.jp\/charlesy\/wp\/wp-content\/uploads\/2020\/02\/Schematic_of_a_replica_exchange_molecular_dynamics_simulation.svg_-300x128.png",300,128,true],"medium_large":["https:\/\/www.ag.kagawa-u.ac.jp\/charlesy\/wp\/wp-content\/uploads\/2020\/02\/Schematic_of_a_replica_exchange_molecular_dynamics_simulation.svg_.png",640,273,false],"large":["https:\/\/www.ag.kagawa-u.ac.jp\/charlesy\/wp\/wp-content\/uploads\/2020\/02\/Schematic_of_a_replica_exchange_molecular_dynamics_simulation.svg_.png",640,273,false],"1536x1536":["https:\/\/www.ag.kagawa-u.ac.jp\/charlesy\/wp\/wp-content\/uploads\/2020\/02\/Schematic_of_a_replica_exchange_molecular_dynamics_simulation.svg_.png",640,273,false],"2048x2048":["https:\/\/www.ag.kagawa-u.ac.jp\/charlesy\/wp\/wp-content\/uploads\/2020\/02\/Schematic_of_a_replica_exchange_molecular_dynamics_simulation.svg_.png",640,273,false],"onepress-blog-small":["https:\/\/www.ag.kagawa-u.ac.jp\/charlesy\/wp\/wp-content\/uploads\/2020\/02\/Schematic_of_a_replica_exchange_molecular_dynamics_simulation.svg_.png",300,128,false],"onepress-small":["https:\/\/www.ag.kagawa-u.ac.jp\/charlesy\/wp\/wp-content\/uploads\/2020\/02\/Schematic_of_a_replica_exchange_molecular_dynamics_simulation.svg_.png",480,205,false],"onepress-medium":["https:\/\/www.ag.kagawa-u.ac.jp\/charlesy\/wp\/wp-content\/uploads\/2020\/02\/Schematic_of_a_replica_exchange_molecular_dynamics_simulation.svg_.png",640,273,false]},"uagb_author_info":{"display_name":"RCY","author_link":"https:\/\/www.ag.kagawa-u.ac.jp\/charlesy\/author\/charlesy\/"},"uagb_comment_info":0,"uagb_excerpt":"\u7814\u7a76\u5ba4\u5185\u30e1\u30e2 \u74b0\u5883: CentOS 7. GROMACS 2 ...","_links":{"self":[{"href":"https:\/\/www.ag.kagawa-u.ac.jp\/charlesy\/wp-json\/wp\/v2\/posts\/1774","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.ag.kagawa-u.ac.jp\/charlesy\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.ag.kagawa-u.ac.jp\/charlesy\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.ag.kagawa-u.ac.jp\/charlesy\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.ag.kagawa-u.ac.jp\/charlesy\/wp-json\/wp\/v2\/comments?post=1774"}],"version-history":[{"count":119,"href":"https:\/\/www.ag.kagawa-u.ac.jp\/charlesy\/wp-json\/wp\/v2\/posts\/1774\/revisions"}],"predecessor-version":[{"id":4068,"href":"https:\/\/www.ag.kagawa-u.ac.jp\/charlesy\/wp-json\/wp\/v2\/posts\/1774\/revisions\/4068"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.ag.kagawa-u.ac.jp\/charlesy\/wp-json\/wp\/v2\/media\/2213"}],"wp:attachment":[{"href":"https:\/\/www.ag.kagawa-u.ac.jp\/charlesy\/wp-json\/wp\/v2\/media?parent=1774"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.ag.kagawa-u.ac.jp\/charlesy\/wp-json\/wp\/v2\/categories?post=1774"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.ag.kagawa-u.ac.jp\/charlesy\/wp-json\/wp\/v2\/tags?post=1774"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}