- Posts: 3
Gromacs 5.0 - Equilibration
- ttubiana
- Topic Author
- Offline
- Fresh Boarder
Less
More
8 years 2 months ago #5382
by ttubiana
Gromacs 5.0 - Equilibration was created by ttubiana
Hello everyone
I know there is already some post about equilibration, but most of them was unanswered or not outdated because of the version 5.0 of gromacs.
So I post a new question :-)
I try to build a good mdp file for the Equilibration step (I will work on a virus Capsid at the end). I tried first on a dimer (~1040 amino acids) and it crashed. There is the error message :
And there is my mdp file (based on Martini mdp file martini_v2.x_new-rf.mdp ):
________________________________________________________________________________________________________________________________
It was runned on a 24 cores system.
Can someone see what's wrong or help me to build a good equilibration step?
Thank you very much :)
Kind regards,
Thibault Tubiana.
I know there is already some post about equilibration, but most of them was unanswered or not outdated because of the version 5.0 of gromacs.
So I post a new question :-)
I try to build a good mdp file for the Equilibration step (I will work on a virus Capsid at the end). I tried first on a dimer (~1040 amino acids) and it crashed. There is the error message :
step 35300, will finish Wed Feb 10 15:50:07 2016vol 0.51 imb F 1%
A list of missing interactions:
Bond of 5669 missing 1
exclusions of 6620 missing 1
-------------------------------------------------------
Program mdrun, VERSION 5.0.7
Source code file: /home/thibault/Downloads/gromacs-5.0.7/src/gromacs/mdlib/domdec_top.c, line: 210
Software inconsistency error:
Some interactions seem to be assigned multiple times
And there is my mdp file (based on Martini mdp file martini_v2.x_new-rf.mdp ):
________________________________________________________________
; Preprocessing
define = -DPOSRES
; Run Control
dt = 0.001 ;1fs
nsteps = 5000000 ;0.001 * 5000000 = 5ns
; Output control
nstxout = 500
nstvout = 0
nstfout = 0
nstlog = 1000
nstenergy = 100
nstxout-compressed = 1000
compressed-x-precision = 1000
; Neighborsearching
cutoff-scheme = Verlet
nstlist = 20
ns_type = grid
pbc = xyz
verlet-buffer-tolerance = 0.005
; Electrostatic
coulombtype = reaction-field
rcoulomb = 1.1
epsilon_r = 15 ; 2.5 (with polarizable water)
epsilon_rf = 0
vdw_type = cutoff
vdw-modifier = Potential-shift-verlet
rvdw = 1.1
;TEMPERATURE PRESSURE
tcoupl = v-rescale
tc-grps = Protein W ION
tau_t = 1.0 1.0 1.0
ref_t = 320 320 320
Pcoupl = Berendsen
Pcoupltype = isotropic
tau_p = 5.0 ;parrinello-rahman is more stable with larger tau-p, DdJ, 20130422
compressibility = 3e-4
ref_p = 1.0
refcoord-scaling = all
gen_vel = yes
gen_temp = 320
gen_seed = -1
;CONSTRAINTS
constraints = none
constraint_algorithm = Lincs
It was runned on a 24 cores system.
Can someone see what's wrong or help me to build a good equilibration step?
Thank you very much :)
Kind regards,
Thibault Tubiana.
Please Log in or Create an account to join the conversation.
- peterkroon
- Offline
- Gold Boarder
Less
More
- Posts: 210
8 years 2 months ago #5385
by peterkroon
Replied by peterkroon on topic Gromacs 5.0 - Equilibration
There's a few things I can spot, but maybe someone else would like to chip in as well.
Firstly, use a larger timestep (i.e. 10fs), unless you have a very good reason not to.
Secondly, don't split your solvent and ions in separate temperature coupling groups ( www.gromacs.org/Documentation/Terminology/Thermostats ).
Thirdly, refcoord-scaling=none should also work, unless you have position restraints.
And lastly, make sure we're not missing anything: how did you create your topology? Did you run an energy minimization prior to this equilibration?
Firstly, use a larger timestep (i.e. 10fs), unless you have a very good reason not to.
Secondly, don't split your solvent and ions in separate temperature coupling groups ( www.gromacs.org/Documentation/Terminology/Thermostats ).
Thirdly, refcoord-scaling=none should also work, unless you have position restraints.
And lastly, make sure we're not missing anything: how did you create your topology? Did you run an energy minimization prior to this equilibration?
Please Log in or Create an account to join the conversation.
- ttubiana
- Topic Author
- Offline
- Fresh Boarder
Less
More
- Posts: 3
8 years 2 months ago #5389
by ttubiana
Replied by ttubiana on topic Gromacs 5.0 - Equilibration
Thank you very much!
It works better now!
I have kept refcoord-scaling=all because I use position restraints.
And yes I've made 2 minimizations (one of the protein in vaccum, then I solvate and neutralize, and I run another minimization, longer). For my topology I use martinyse.py :)
I also reduce the number of core to 16. And it sounds like with this protocol more than 32 cores is not allowed...
Kind Regards,
Thibault.
It works better now!
I have kept refcoord-scaling=all because I use position restraints.
And yes I've made 2 minimizations (one of the protein in vaccum, then I solvate and neutralize, and I run another minimization, longer). For my topology I use martinyse.py :)
I also reduce the number of core to 16. And it sounds like with this protocol more than 32 cores is not allowed...
Kind Regards,
Thibault.
Please Log in or Create an account to join the conversation.
- peterkroon
- Offline
- Gold Boarder
Less
More
- Posts: 210
8 years 2 months ago #5390
by peterkroon
Replied by peterkroon on topic Gromacs 5.0 - Equilibration
No problem.
The number of cores is usually limited by the system size: a larger systems allows for more parallelization.
If you run into further troubles, what version of Martinize and Martini are you using?
The number of cores is usually limited by the system size: a larger systems allows for more parallelization.
If you run into further troubles, what version of Martinize and Martini are you using?
Please Log in or Create an account to join the conversation.
- siti
- Offline
- Fresh Boarder
Less
More
- Posts: 3
8 years 2 months ago #5394
by siti
Replied by siti on topic Gromacs 5.0 - Equilibration
Hi,
I would like to know, if there is updated page to get the new mdp (minimization.mdp, equlibration.mdp) files to run protein in DPPC bilayer using the martini ff for the Gromacs 5.
The files that I downloaded from this page:
www.cgmartini.nl/index.php/tutorials-gen...ntroduction/proteins
are not updated for the Gromacs 5.0.
Thank you for your help
I would like to know, if there is updated page to get the new mdp (minimization.mdp, equlibration.mdp) files to run protein in DPPC bilayer using the martini ff for the Gromacs 5.
The files that I downloaded from this page:
www.cgmartini.nl/index.php/tutorials-gen...ntroduction/proteins
are not updated for the Gromacs 5.0.
Thank you for your help
Please Log in or Create an account to join the conversation.
- peterkroon
- Offline
- Gold Boarder
Less
More
- Posts: 210
8 years 1 month ago #5405
by peterkroon
Replied by peterkroon on topic Gromacs 5.0 - Equilibration
Try
cgmartini.nl/index.php/mdp
, and particularly the new-rf. These parameters come from
www.sciencedirect.com/science/article/pii/S0010465515003628
.
I'll see if I can get someone to update the tutorial(s).
I'll see if I can get someone to update the tutorial(s).
Please Log in or Create an account to join the conversation.
- yogi@martini
- Offline
- Junior Boarder
Less
More
- Posts: 31
7 years 3 months ago #6980
by yogi@martini
Replied by yogi@martini on topic Gromacs 5.0 - Equilibration
Hello friend,
The minimization.mdp for Gromacs 5 is still not updated on the website.
Can someone paste the content of the mdp file here itself so that we can use.
Thanks
The minimization.mdp for Gromacs 5 is still not updated on the website.
Can someone paste the content of the mdp file here itself so that we can use.
Thanks
Please Log in or Create an account to join the conversation.
- riccardo
- Offline
- Platinum Boarder
7 years 2 months ago - 7 years 2 months ago #7071
by riccardo
Replied by riccardo on topic Gromacs 5.0 - Equilibration
Just pick your favourite set of parameters you find at:
www.cgmartini.nl/index.php/force-field-p...ers/input-parameters
(please check also the associated reference)
and adapt to get an minimization mdp set. Basically you'll need to change the following couple of lines:
integrator = steep
nsteps = 20000
www.cgmartini.nl/index.php/force-field-p...ers/input-parameters
(please check also the associated reference)
and adapt to get an minimization mdp set. Basically you'll need to change the following couple of lines:
integrator = steep
nsteps = 20000
Last edit: 7 years 2 months ago by riccardo.
Please Log in or Create an account to join the conversation.
- yogi@martini
- Offline
- Junior Boarder
Less
More
- Posts: 31
7 years 2 months ago #7120
by yogi@martini
Replied by yogi@martini on topic Gromacs 5.0 - Equilibration
Thanks Riccardo!
Please Log in or Create an account to join the conversation.
Time to create page: 0.117 seconds