1GMX-MAKE_EDI(1) GROMACS GMX-MAKE_EDI(1)
2
6 gmx-make_edi - Generate input files for essential dynamics sampling
7
9 gmx make_edi [-f [<.trr/.cpt/...>]] [-eig [<.xvg>]]
10 [-s [<.tpr/.gro/...>]] [-n [<.ndx>]]
11 [-tar [<.gro/.g96/...>]] [-ori [<.gro/.g96/...>]]
12 [-o [<.edi>]] [-xvg <enum>] [-mon <string>]
13 [-linfix <string>] [-linacc <string>] [-radfix <string>]
14 [-radacc <string>] [-radcon <string>] [-flood <string>]
15 [-outfrq <int>] [-slope <real>] [-linstep <string>]
16 [-accdir <string>] [-radstep <real>] [-maxedsteps <int>]
17 [-eqsteps <int>] [-deltaF0 <real>] [-deltaF <real>]
18 [-tau <real>] [-Eflnull <real>] [-T <real>]
19 [-alpha <real>] [-[no]restrain] [-[no]hessian]
20 [-[no]harmonic] [-constF <string>]
21
23 gmx make_edi generates an essential dynamics (ED) sampling input file
24 to be used with mdrun based on eigenvectors of a covariance matrix (gmx
25 covar) or from a normal modes analysis (gmx nmeig). ED sampling can be
26 used to manipulate the position along collective coordinates (eigenvec‐
27 tors) of (biological) macromolecules during a simulation. Particularly,
28 it may be used to enhance the sampling efficiency of MD simulations by
29 stimulating the system to explore new regions along these collective
30 coordinates. A number of different algorithms are implemented to drive
31 the system along the eigenvectors (-linfix, -linacc, -radfix, -radacc,
32 -radcon), to keep the position along a certain (set of) coordinate(s)
33 fixed (-linfix), or to only monitor the projections of the positions
34 onto these coordinates (-mon).
35 References:
37 A. Amadei, A.B.M. Linssen, B.L. de Groot, D.M.F. van Aalten and H.J.C.
39 Berendsen; An efficient method for sampling the essential subspace of
40 proteins., J. Biomol. Struct. Dyn. 13:615-626 (1996)
41 B.L. de Groot, A. Amadei, D.M.F. van Aalten and H.J.C. Berendsen;
43 Towards an exhaustive sampling of the configurational spaces of the two
44 forms of the peptide hormone guanylin, J. Biomol. Struct. Dyn. 13 :
45 741-751 (1996)
46 B.L. de Groot, A.Amadei, R.M. Scheek, N.A.J. van Nuland and H.J.C.
48 Berendsen; An extended sampling of the configurational space of HPr
49 from E. coli Proteins: Struct. Funct. Gen. 26: 314-322 (1996)
50 You will be prompted for one or more index groups that correspond to
52 the eigenvectors, reference structure, target positions, etc.
53 -mon: monitor projections of the coordinates onto selected eigenvec‐
55 tors.
56 -linfix: perform fixed-step linear expansion along selected eigenvec‐
58 tors.
59 -linacc: perform acceptance linear expansion along selected eigenvec‐
61 tors. (steps in the desired directions will be accepted, others will
62 be rejected).
63 -radfix: perform fixed-step radius expansion along selected eigenvec‐
65 tors.
66 -radacc: perform acceptance radius expansion along selected eigenvec‐
68 tors. (steps in the desired direction will be accepted, others will be
69 rejected). Note: by default the starting MD structure will be taken as
70 origin of the first expansion cycle for radius expansion. If -ori is
71 specified, you will be able to read in a structure file that defines an
72 external origin.
73 -radcon: perform acceptance radius contraction along selected eigenvec‐
75 tors towards a target structure specified with -tar.
76 NOTE: each eigenvector can be selected only once.
78 -outfrq: frequency (in steps) of writing out projections etc. to .xvg
80 file
81 -slope: minimal slope in acceptance radius expansion. A new expansion
83 cycle will be started if the spontaneous increase of the radius (in
84 nm/step) is less than the value specified.
85 -maxedsteps: maximum number of steps per cycle in radius expansion
87 before a new cycle is started.
88 Note on the parallel implementation: since ED sampling is a ‘global’
90 thing (collective coordinates etc.), at least on the ‘protein’ side, ED
91 sampling is not very parallel-friendly from an implementation point of
92 view. Because parallel ED requires some extra communication, expect the
93 performance to be lower as in a free MD simulation, especially on a
94 large number of ranks and/or when the ED group contains a lot of atoms.
95 Please also note that if your ED group contains more than a single pro‐
97 tein, then the .tpr file must contain the correct PBC representation of
98 the ED group. Take a look on the initial RMSD from the reference
99 structure, which is printed out at the start of the simulation; if this
100 is much higher than expected, one of the ED molecules might be shifted
101 by a box vector.
102 All ED-related output of mdrun (specify with -eo) is written to a .xvg
104 file as a function of time in intervals of OUTFRQ steps.
105 Note that you can impose multiple ED constraints and flooding poten‐
107 tials in a single simulation (on different molecules) if several .edi
108 files were concatenated first. The constraints are applied in the order
109 they appear in the .edi file. Depending on what was specified in the
110 .edi input file, the output file contains for each ED dataset
111 · the RMSD of the fitted molecule to the reference structure (for
113 atoms involved in fitting prior to calculating the ED constraints)
114 · projections of the positions onto selected eigenvectors
116 FLOODING:
118 with -flood, you can specify which eigenvectors are used to compute a
120 flooding potential, which will lead to extra forces expelling the
121 structure out of the region described by the covariance matrix. If you
122 switch -restrain the potential is inverted and the structure is kept in
123 that region.
124 The origin is normally the average structure stored in the eigvec.trr
126 file. It can be changed with -ori to an arbitrary position in configu‐
127 ration space. With -tau, -deltaF0, and -Eflnull you control the flood‐
128 ing behaviour. Efl is the flooding strength, it is updated according
129 to the rule of adaptive flooding. Tau is the time constant of adaptive
130 flooding, high tau means slow adaption (i.e. growth). DeltaF0 is the
131 flooding strength you want to reach after tau ps of simulation. To use
132 constant Efl set -tau to zero.
133 -alpha is a fudge parameter to control the width of the flooding poten‐
135 tial. A value of 2 has been found to give good results for most stan‐
136 dard cases in flooding of proteins. alpha basically accounts for
137 incomplete sampling, if you sampled further the width of the ensemble
138 would increase, this is mimicked by alpha > 1. For restraining, alpha
139 < 1 can give you smaller width in the restraining potential.
140 RESTART and FLOODING: If you want to restart a crashed flooding simula‐
142 tion please find the values deltaF and Efl in the output file and manu‐
143 ally put them into the .edi file under DELTA_F0 and EFL_NULL.
144
146 Options to specify input files:
147 -f [<.trr/.cpt/…>] (eigenvec.trr)
149 Full precision trajectory: trr cpt tng
150 -eig [<.xvg>] (eigenval.xvg) (Optional)
152 xvgr/xmgr file
153 -s [<.tpr/.gro/…>] (topol.tpr)
155 Structure+mass(db): tpr gro g96 pdb brk ent
156 -n [<.ndx>] (index.ndx) (Optional)
158 Index file
159 -tar [<.gro/.g96/…>] (target.gro) (Optional)
161 Structure file: gro g96 pdb brk ent esp tpr
162 -ori [<.gro/.g96/…>] (origin.gro) (Optional)
164 Structure file: gro g96 pdb brk ent esp tpr
165 Options to specify output files:
167 -o [<.edi>] (sam.edi)
169 ED sampling input
170 Other options:
172 -xvg <enum> (xmgrace)
174 xvg plot formatting: xmgrace, xmgr, none
175 -mon <string>
177 Indices of eigenvectors for projections of x (e.g. 1,2-5,9) or
178 1-100:10 means 1 11 21 31 … 91
179 -linfix <string>
181 Indices of eigenvectors for fixed increment linear sampling
182 -linacc <string>
184 Indices of eigenvectors for acceptance linear sampling
185 -radfix <string>
187 Indices of eigenvectors for fixed increment radius expansion
188 -radacc <string>
190 Indices of eigenvectors for acceptance radius expansion
191 -radcon <string>
193 Indices of eigenvectors for acceptance radius contraction
194 -flood <string>
196 Indices of eigenvectors for flooding
197 -outfrq <int> (100)
199 Frequency (in steps) of writing output in .xvg file
200 -slope <real> (0)
202 Minimal slope in acceptance radius expansion
203 -linstep <string>
205 Stepsizes (nm/step) for fixed increment linear sampling (put in
206 quotes! “1.0 2.3 5.1 -3.1”)
207 -accdir <string>
209 Directions for acceptance linear sampling - only sign counts!
210 (put in quotes! “-1 +1 -1.1”)
211 -radstep <real> (0)
213 Stepsize (nm/step) for fixed increment radius expansion
214 -maxedsteps <int> (0)
216 Maximum number of steps per cycle
217 -eqsteps <int> (0)
219 Number of steps to run without any perturbations
220 -deltaF0 <real> (150)
222 Target destabilization energy for flooding
223 -deltaF <real> (0)
225 Start deltaF with this parameter - default 0, nonzero values
226 only needed for restart
227 -tau <real> (0.1)
229 Coupling constant for adaption of flooding strength according to
230 deltaF0, 0 = infinity i.e. constant flooding strength
231 -Eflnull <real> (0)
233 The starting value of the flooding strength. The flooding
234 strength is updated according to the adaptive flooding scheme.
235 For a constant flooding strength use -tau 0.
236 -T <real> (300)
238 T is temperature, the value is needed if you want to do flooding
239 -alpha <real> (1)
241 Scale width of gaussian flooding potential with alpha^2
242 -[no]restrain (no)
244 Use the flooding potential with inverted sign -> effects as
245 quasiharmonic restraining potential
246 -[no]hessian (no)
248 The eigenvectors and eigenvalues are from a Hessian matrix
249 -[no]harmonic (no)
251 The eigenvalues are interpreted as spring constant
252 -constF <string>
254 Constant force flooding: manually set the forces for the eigen‐
255 vectors selected with -flood (put in quotes! “1.0 2.3 5.1
256 -3.1”). No other flooding parameters are needed when specifying
257 the forces directly.
258
260 gmx(1)
261 More information about GROMACS is available at <‐
263 http://www.gromacs.org/>.
264
266 2020, GROMACS development team
2672019.6 Feb 28, 2020 GMX-MAKE_EDI(1)