Naproxen parameters

ric.concu
Topic Author
Offline
Fresh Boarder

6 years 3 months ago #7491 by ric.concu

Naproxen parameters was created by ric.concu

Dear all,
I am writing in this forum looking for any kind of advice or tips that could help me in setting up the parameters for the parametrization of the Naproxen. I followed the tutorial on the home page however, I am really unable to find a stable configuration for this molecule. I am able to perform some simulation with the molecule alone. Although it seems working fine when I try to simulate in a box with water the simulation didn't go properly. This is my itp and how I have been defining the beads. I am using classical mdp from the martini tutorials.
Thank you in advance

Warning: Spoiler! [ Click to expand ]

Please Log in or Create an account to join the conversation.

bart
Away
Admin

6 years 3 months ago #7492 by bart

Replied by bart on topic Naproxen parameters

I don't see anything too strange at first glance. What do you mean with finding a stable configuration? What is the timestep you are using (I would start with something like 10-20 fs after equilibration with a timestep around 2 fs)? I think the problem is most likely not in the itp of this molecule, but the settings you use for running it.

Cheers,

Bart

Please Log in or Create an account to join the conversation.

ric.concu
Topic Author
Offline
Fresh Boarder

6 years 3 months ago #7495 by ric.concu

Replied by ric.concu on topic Naproxen parameters

Dear Bart,
thank you for your reply.
I ran a short simulation with Nap in water and the simulation seems to be okay, however, the final gro shows that the system exploded. The mdp options I am using are listed here below

integrator = md
dt = 0.0002
nsteps = 100000
nstcomm = 100
nstxout = 0
nstvout = 0
nstfout = 0
nstlog = 1000
nstenergy = 100
nstxout-compressed = 1000
compressed-x-precision = 100
cutoff-scheme = Verlet
nstlist = 10
ns_type = grid
pbc = xyz
rlist = 1.4
coulombtype = cutoff
rcoulomb_switch = 0.0
rcoulomb = 1.2
epsilon_r = 15 ; 2.5 (with polarizable water)
vdw_type = cutoff
rvdw_switch = 0.9
rvdw = 1.2
tcoupl = v-rescale
tc-grps = system
tau_t = 1.0
ref_t = 320
Pcoupl = parrinello-rahman
Pcoupltype = isotropic
tau_p = 12.0 ;parrinello-rahman is more stable with larger tau-p, DdJ, 20130422
compressibility = 3e-4
ref_p = 1.0

gen_vel = no
gen_temp = 320
gen_seed = 473529
constraints = none
constraint_algorithm = Lincs

Thank you for helping,
Riccardo

Please Log in or Create an account to join the conversation.

riccardo
Offline
Platinum Boarder

6 years 3 months ago - 6 years 3 months ago #7496 by riccardo

Replied by riccardo on topic Naproxen parameters

You should not need such a small timestep.
Usually one first minimizes the system (with steepest descent), then run an NVT equilibration (using a Berendsen thermostat), then NPT eq (both Berendsen thermostat and barostats) and finally a run with settings similar to the ones you used (but with at least some 10 fs as timestep). Also what Bart mentioned is that , in case of instabilities, you could use a dt of 2 fs in the NVT and NPT equilibrations.

Also, you say you don't get any error message during the simulation / your simulation doesn't actually crash?
What makes you think the system has exploded in the last snapshot (your molecules' beads are very far from each other?)?

Last edit: 6 years 3 months ago by riccardo.

Please Log in or Create an account to join the conversation.

ric.concu
Topic Author
Offline
Fresh Boarder

6 years 3 months ago #7497 by ric.concu

Replied by ric.concu on topic Naproxen parameters

Dear,
I previously ran an Energy minimization with steepest descent and everytihng seems ok. Then I have been trying a NVT equilibration with Berendsen thermostat but it stops due too many LINCS warning. Due to this I tried a relaxation step (as in the Martini tutorial) but also this run miserably fails for LINCS warning. I am supposing that the system in the npt is blowing up because the the gro file looks like:
1NAP C1 11154.666 988.2201255.266 0.4971 1.6701 -4.3845
1NAP C2 21154.544 988.4531255.170 -8.4040 -1.9099 -2.0947
1NAP C3 31154.695 988.5661255.282 -5.2645 -3.8749 -3.9720
1NAP C4 41154.506 988.6561255.268 0.5445 10.6017 3.8307
1NAP C5 51154.391 988.6511255.130 5.0321 -3.1450 -0.6800
1NAP C6 61154.620 988.3591255.050 7.2961 -4.3438 9.9304
2NAP C1 71382.3181276.6201400.228 -2.6060 -0.0474 10.4751
2NAP C2 81382.0891276.7811400.218 -2.5068 -1.6939 -6.3560

Please Log in or Create an account to join the conversation.

peterkroon
Offline
Gold Boarder

6 years 3 months ago - 6 years 3 months ago #7498 by peterkroon

Replied by peterkroon on topic Naproxen parameters

Your NAP does not define constraints. What molecule is giving LINCS warnings? Are you using polarizable water?

Last edit: 6 years 3 months ago by peterkroon.

Please Log in or Create an account to join the conversation.

ric.concu
Topic Author
Offline
Fresh Boarder

6 years 3 months ago #7499 by ric.concu

Replied by ric.concu on topic Naproxen parameters

I also tried using constraints on bonds however the result is the same. I am using the water and WF form the Martiniv2.2

Please Log in or Create an account to join the conversation.

peterkroon
Offline
Gold Boarder

6 years 3 months ago #7500 by peterkroon

Replied by peterkroon on topic Naproxen parameters

You said your run crashed due to LINCS warnings. This is impossible with the mdp and parameters you posted earlier. What parameters give what errors?

Please Log in or Create an account to join the conversation.

ric.concu
Topic Author
Offline
Fresh Boarder

6 years 3 months ago #7501 by ric.concu

Replied by ric.concu on topic Naproxen parameters

All the LINCS WARNING are like this one:
Step 81, time 0.162 (ps) LINCS WARNING
relative constraint deviation after LINCS:
rms 74139.070312, max 174495.281250 (between atoms 42 and 38)
bonds that rotated more than 30 degrees:
atom 1 atom 2 angle previous, current, constraint length

Please Log in or Create an account to join the conversation.

peterkroon
Offline
Gold Boarder

6 years 3 months ago #7502 by peterkroon

Replied by peterkroon on topic Naproxen parameters

This is *impossible* with the itp and mdp options you provided.

Please also post your complete .top file.

Please Log in or Create an account to join the conversation.

ric.concu
Topic Author
Offline
Fresh Boarder

6 years 3 months ago #7503 by ric.concu

Replied by ric.concu on topic Naproxen parameters

Here it comes:
#include "martini_v2.2.itp"
#include "napCG_opt.itp"
#include "ions.itp"

[ system ]
NAP water

[ molecules ]
NAP 10
NA 10
W 5000
WF 500

Thank you all for your kind help

Please Log in or Create an account to join the conversation.

peterkroon
Offline
Gold Boarder

6 years 3 months ago #7504 by peterkroon

Replied by peterkroon on topic Naproxen parameters

You're not simulating what you think you're simulating. There's no constraints in your naproxene itp (first post), nor in your mdp file, yet you're getting errors related to constraints.

Please Log in or Create an account to join the conversation.

ric.concu
Topic Author
Offline
Fresh Boarder

6 years 3 months ago #7505 by ric.concu

Replied by ric.concu on topic Naproxen parameters

Even with constrints on the NVT equilibration is stopping due to LINCS warning. These are my mdp and itp entries:

integrator = md
dt = 0.002
nsteps = 500000
nstxout = 5000
nstvout = 5000
nstlog = 5000
nstenergy = 300
nstxout-compressed = 250
compressed-x-grps = System
energygrps = System
nstlist = 20
ns-type = grid
rlist = 1.0
coulombtype = cut-off
rcoulomb = 1.0
rvdw = 1.0
tcoupl = Berendsen
tc-grps = System
tau-t = 0.2
ref-t = 300
Pcoupl = Berendsen
tau-p = 1.0
compressibility = 4.5e-5
ref-p = 1.0
gen-vel = yes
gen-temp = 300
gen-seed = 173529
constraints = all-bonds

[ moleculetype ]
; molname nrexcl
NAP 1

[ atoms ]
;id type resnr residu atom cgnr charge
1 Qd 1 NAP C1 1 -1
2 SC5 1 NAP C2 2 0
3 SC5 1 NAP C3 3 0
4 SC5 1 NAP C4 4 0
5 SC4 1 NAP C5 5 0
6 SC5 1 NAP C6 6 0

;[bonds]
;; i j funct length force.c.
;1 2 1 0.30 12500
;2 3 1 0.30 12500
;3 4 1 0.31 12500
;4 5 1 0.45 12500
;5 6 1 0.28 12500
;6 2 1 0.28 12500

[constraints]
; i j funct length
1 2 1 0.28
2 3 1 0.22
3 4 1 0.21
4 5 1 0.18
5 6 1 0.38
6 2 1 0.17

[angles]
; i j k funct angle force.c.
1 2 3 2 120.0 250
2 3 4 2 150.0 250
3 4 5 2 65.0 250
4 5 6 2 110.0 250
5 6 2 2 110.0 250
6 2 3 2 104.0 250
1 2 6 2 140.0 250

[dihedrals]
; i j k l funct angle force.c.
;1 2 3 4 3
;2 3 4 5 3
;3 4 5 6 3
;4 5 6 2 3
;1 2 6 5 3

Please Log in or Create an account to join the conversation.

riccardo
Offline
Platinum Boarder

6 years 3 months ago #7507 by riccardo

Replied by riccardo on topic Naproxen parameters

I repeat as well what Peter was saying: with the first topology you posted (which contained no constraints) and the first mdp you posted (which contained "constraints = none") it is *not* possible that you get LINCS warnings. You get LINCS warnings in the case the constraints solver (i.e., the LINCS algorithm) struggles solving the constraints. But if you have no constraints (which is the case if you use the first itp and mdp you posted) then you *cannot* be getting LINCS warnings.

My guess is that you were grompping (gmx grompp) the files multiple times in the same folder trying to make your system work and started to lose track of which tpr was produced by which set of top, gro and mdp. Can you try a run (in a new folder) with the first itp (the one without constraints) and mdp's where you set 'constraints = none'?

Please Log in or Create an account to join the conversation.

ric.concu
Topic Author
Offline
Fresh Boarder

6 years 3 months ago #7509 by ric.concu

Replied by ric.concu on topic Naproxen parameters

You both were right, however now the simulation stops with the segmentation fault core dumped error.

Please Log in or Create an account to join the conversation.

riccardo
Offline
Platinum Boarder

6 years 3 months ago #7510 by riccardo

Replied by riccardo on topic Naproxen parameters

Alright. Does the minimization converge? Do you get the seg fault as soon as the NVT starts?

Please Log in or Create an account to join the conversation.

Pim
Offline
Expert Boarder

6 years 3 months ago #7511 by Pim

Replied by Pim on topic Naproxen parameters

Just weighing in; your mapping in combination with your bonds and angles as you define them are not so great and are sensitive to crashing I think, with the naphthalene moiety modeled as 5 beads in a circle, no exclusions and no improper dihedrals to keep it flat. We typically would map naphthalene to probably 4 beads, or 4 beads and a virtual site, such as in genistein (SI of pubs.acs.org/doi/full/10.1021/cb500086e ).

Please Log in or Create an account to join the conversation.

ric.concu
Topic Author
Offline
Fresh Boarder

6 years 3 months ago #7513 by ric.concu

Replied by ric.concu on topic Naproxen parameters

The EM yes converge and the simulation in the NVT ensemble stops as soon as it starts with the segmentation fault error

Please Log in or Create an account to join the conversation.

ric.concu
Topic Author
Offline
Fresh Boarder

6 years 3 months ago #7514 by ric.concu

Replied by ric.concu on topic Naproxen parameters

Dear Pim, thank you for your advice. I will try to map the naproxen using a 4 beads mapping for the napthalene moiety.

Please Log in or Create an account to join the conversation.

Pim
Offline
Expert Boarder

6 years 3 months ago #7515 by Pim

Replied by Pim on topic Naproxen parameters

Maybe riccardo can help you even further with a better fitting topology+mapping, he's the aromatic ring expert :).

Please Log in or Create an account to join the conversation.

Time to create page: 0.135 seconds

General Purpose Coarse-Grained Force Field

Nav view search

Navigation

Naproxen parameters