Today I have a question about my solvation. After immersing in a droplet of water, I try to equilbrium, but there is a abnormal termination when heating. This is the code:
DYNA DYN: Step Time TOTEner TOTKe ENERgy TEMPerature DYNA PROP: GRMS HFCTote HFCKe EHFCor VIRKe DYNA INTERN: BONDs ANGLes UREY-b DIHEdrals IMPRopers DYNA EXTERN: VDWaals ELEC HBONds ASP USER DYNA MISCEL: SBOUnd TSM PRMS PANGle other DYNA PRESS: VIRE VIRI PRESSE PRESSI VOLUme ---------- --------- --------- --------- --------- --------- DYNA> 0 0.00000-299487.95558 0.00000-299487.95558 0.00000 DYNA PROP> 28.72222-299327.78729 480.50490 160.16830 -13858.10193 DYNA INTERN> 0.00000 922.19829 207.86836 1105.91786 46.22287 DYNA EXTERN> 10501.32156-312122.45716 0.00000 0.00000 0.00000 DYNA MISCEL> -149.02736 0.00000 0.00000 0.00000 0.00000 DYNA PRESS> 210133.59958-200894.86496 0.00000 0.00000 0.00000 ---------- --------- --------- --------- --------- --------- UPDECI: Nonbond update at step 1 UPDECI: Nonbond update at step 6 UPDECI: Nonbond update at step 10 UPDECI: Nonbond update at step 13 UPDECI: Nonbond update at step 14 WRIDYN: RESTart file was written at step 15
Writing RESTART FILE with previous and current coordinates, which may be read by: READ COOR DYNR .... (see io.doc). NOTE!! THIS FILE C A N N O T BE USED TO RESTART A RUN!!!
***** LEVEL -2 WARNING FROM ***** ***** ENERGY CHANGE TOLERANCE EXCEEDED ****************************************** BOMLEV ( -1) IS REACHED - TERMINATING. WRNLEV IS 0
ABNORMAL TERMINATION MOST SEVERE WARNING WAS AT LEVEL -2
$$$$$ JOB ACCOUNTING INFORMATION $$$$$ ELAPSED TIME: 20.79 MINUTES CPU TIME: 20.70 MINUTES
So how can I deal with it? help would be greatly appreciated wenhao
One possibility is that the structure has not been adequately minimized after solvation. There could also be some issues with your dynamics input script. One thing to note is that the last error or warning message may be a symptom of an earlier problem.
Thank you so much. Now I still try my best to deal with this problem. But I dont know what is your meaning about "an earlier problem"? To be honest,after my first minimization, the first residue and the final resdue were abnormal(bonding of unrelated atoms that needn‘t bond). But I used an order:define heavy sele all .and. .not. ( bynu 1:28 .or. bynu 12547:12569 .or. resname TIP3 .or. hydrogen) end.Then the minimization was fine and the mini file were used as input file for solvation.
Thank you very much ,really. I should tell you about the step of minimization : the out PSF file of minimization before solvation is normal(displaying no bond), but when I open the out PDB-minimization using VMD,the image is abnormal for the first residue and last residue. So my teacher told me to use the order. I will cannot deal with the abnormal bonds without the order, sorry, I am a new learner for Charmm. I really thank you for your reply and your time.There are log files in the attachment. Thank you again and wish you happy every day.
VMD will not display bonds correctly unless you read a PSF before coordinates; for CHARMM, using the COOR file (VMD .cor) is recommended, using PDB is not
Input files are also quite informative, and a lot easier to read than lengthy log files. You should strive to post a very brief input file (and the corresponding log file) that shows the alleged bad behavior. Do not use wrnlev, prnlev or bomlev commands. Your logfiles are so cluttered with mostly uninteresting information that they are unreadable. A "define" command does not in itself do anything. The PSF is not changed by coordinate manipulations such as minimization. And please always use the most recent version of CHARMM.
Lennart Nilsson Karolinska Institutet Stockholm, Sweden
Yes，teacher. I definitely used the COOR file for Charmm. I really cannot deal with this abnormal termination, even if I change the inp file for equilbrium:shake bonh tolerance 1.0e-6 parameter (change the number). How can I do next?
Thank you for your reply and suggestions. I still cannot deal with this complicated termination in heating 300K. There are input files and corresponding log files in the attachment. Thank you again and wish you happy every day.
Please stop using PRNLEV 10 (or 15) for these output logs if you want us to look through them; use the default (PRNLEV 5), or PRNLEV 6 at most.
The CHARMM version is several years out of date.
You also appear to be using out of date, modified parameters, which could be one of the many problems in this set of files. The extraneous character warning indicates an issue with how the file edited at some point. You should resolve the errors that occur during parameter reading.
Code
CHARMM> ! Parameter files
CHARMM> open read unit 2 card name /home/dengwenhao/5ymr/par_all_isethionate.prm
VOPEN> Attempting to open::/home/dengwenhao/5ymr/par_all_isethionate.prm::
OPNLGU> Unit 2 opened for READONLY access to /home/dengwenhao/5ymr/par_all_isethionate.prm
CHARMM> read para card unit 2
PARAMETER FILE BEING READ FROM UNIT 2
TITLE> *>>>>>> COMBINED CHARMM ALL-HYDROGEN PARAMETER FILE FOR <<<<<<<<<
TITLE> *>>>>>>> CHARMM22 PROTEINS AND CHARMM27 NUCLEIC ACIDS <<<<<<<<<<
TITLE> *FROM
TITLE> *>>>> CHARMM22 ALL-HYDROGEN PARAMETER FILE FOR PROTEINS <<<<<<<<
TITLE> *>>>>>>>>>>>>>>>>>>>>>>> AUGUST 1999 <<<<<<<<<<<<<<<<<<<<<<<<<<<
TITLE> *>>>>>>> DIRECT COMMENTS TO ALEXANDER D. MACKERELL JR. <<<<<<<<<
TITLE> *>>>>>> 410-706-7442 OR EMAIL: ALEX,MMIRIS.AB.UMD.EDU <<<<<<<<<
TITLE> *AND
TITLE> * \\\\ CHARMM27 ALL-HYDROGEN NUCLEIC ACID PARAMETER FILE ////
TITLE> * \\\\\\\\\\\\\\\\\\ DEVELOPMENTAL ////////////////////////
TITLE> * ALEXANDER D. MACKERELL JR. AND NICOLAS FOLOPPE
TITLE> * AUGUST 1999
TITLE> * ALL COMMENTS TO ADM JR. EMAIL: ALEX,MMIRIS.AB.UMD.EDU
TITLE> * TELEPHONE: 410-706-7442
TITLE> *
**** Warning **** The following extraneous characters
were found while command processing in PARRDR
£°
PARRDR> WARNING: ATOM FOR NBOND OSL DOESNT EXIST
PARRDR> ALL ANGLES HAVE POSITIVE MINIMA
PARRDR> Error: Repeated BOND parameter ( 19992): SF FE
PARRDR> Error: Repeated BOND parameter ( 14901): ON2 CT2
PARRDR> Error: Repeated BOND parameter ( 11579): NN2 HN2
PARRDR> Error: Repeated BOND parameter ( 4269): FE SS
PARRDR> Error: Repeated BOND parameter ( 4246): FE NPH
PARRDR> Error: Repeated BOND parameter ( 4238): FE NR2
PARRDR> Error: Repeated BOND parameter ( 4223): FE CM
PARRDR> Error: Repeated BOND parameter ( 4222): FE CPM
The initial minimization energies indicate possible issues with the model building; very high VDW energy. You should also be using SHAKE BONH PARAM at this point, and with the default 1.e-9 tolerance; the same applies to the post-solvation minimization.
Why are you using CONS FIX for all the heavy atoms? This is highly inadvisable, and means you aren't really doing much minimization.
This is a good example of the wrong way to condition an initial model to prepare it for dynamics.
Thanks，my teacher. Thank you for your suggestions and your time. And this time I tried again and again.
I used SHAKE BONH PARAM at initial minimization before solvation, but I found the energy was higher than previous(not using SHAKE BONH PARAM, shake bonh tolerance 1.0e-9 ). So which one is better?
When I used SHAKE BONH PARAM at the post-solvation minimization( set the number 1.0e-9, not previously1.0e-6), the abnormal termination still happened. I felt disappointed very much.I just want to try again and I deleted "shake bonh tolerance 1.0e-9 parameter shake bonh parameter" in inp for equilbirum, then it worked and the abnormal termination didn't happen. So I want to ask: Can I delete this script "shake bonh tolerance 1.0e-9 parameter shake bonh parameter" ? I am afraid for this action. Is this deleting operation harmful for next steps?
Thank you again for your time, my teacher, thank you!
Thanks，my teacher. Thank you for your suggestions and your time. And this time I tried again and again.
I used SHAKE BONH PARAM at initial minimization before solvation, but I found the energy was higher than previous(not using SHAKE BONH PARAM, shake bonh tolerance 1.0e-9 ). So which one is better?
When I used SHAKE BONH PARAM at the post-solvation minimization( set the number 1.0e-9, not previously1.0e-6), the abnormal termination still happened. I felt disappointed very much.I just want to try again and I deleted "shake bonh tolerance 1.0e-9 parameter shake bonh parameter" in inp for equilbirum, then it worked and the abnormal termination didn't happen. So I want to ask: Can I delete this script "shake bonh tolerance 1.0e-9 parameter shake bonh parameter" ? I am afraid for this action. Is this deleting operation harmful for next steps?
Thank you again for your time, my teacher, thank you!
You must use SHAKE BONH PARAM throughout; the errors are caused by other problems, such as the use of CONS FIX instead of harmonic restraints, a possibly corrupted parameter file, and maybe an initial model that isn't very good.
Verify the parameter file, perhaps using a simpler model.
Revise the minimizations to use CONS HARM instead of CONS FIX, and do not apply it to all heavy atoms; just the backbone atoms is likely a better choice. The best choices depend on the source of the model; I would treat a crystal structure differently that one constructed from a software package.
Ok，thanks. Though I don't know the reason why we must use SHAKE BONH PARAM throughout, but I believe you are right. At the same time, maybe I know where I should change, I almost forget ask : how do I change the script to achieve the purpose of applying "CONS HARM"(your suggestion) to the backbone atoms instead of all heavy atoms.
This is my old script: !Make sure no constrains exist cons fix sele none end
define heavy sele all .and. .not. ( resname TIP3 .or. hydrogen) show end
!Constrain needed locations cons fix sele heavy end
How do I change my old script, my teacher?
Thank you again for your valuable suggestions. Thank you!
The CHARMM force field was designed and tested using the SHAKE BONH constraint. The water model in particular is rigid, and requires SHAKE to keep it that way. The PARAM keyword ensures that X-H bonds are constrained to their parameter set values.
Please read the documentation (cons), and look at some the posts in the Script Archive to get some idea of how this is supposed to be done.
Using CONS FIX on all heavy atoms and leaving only the H atoms free to move is an extremely bad idea, indicating either you have received very bad advice, or have not done the proper amount of study yourself before attempting to use this complex program. Make some effort, and show that you have done so.
