Hi Kry,
Note if minimization does not work on the CPU there is no way it would work on the GPU. The reason the CPU is recommended for minimization is because it uses a full double precision representation of the forces while the GPU code uses fixed precision integers. For MD this is fine because for a well behaved system the forces in internal units should never exceed the fixed precision maximum. However, when minimizing one can have bad contacts and very high initial forces. With the GPU code these large forces get truncated which can confuse the minimizer. Hence the GPU code is only suitable for minimizing structures that do not contain any bad contacts or other high strain situations. For this reason it is always safest to just minimize with the CPU code.
Hope that helps,
All the best
Ross
> On Sep 11, 2017, at 11:43 AM, Lizelle Lubbe <LBBLIZ002.myuct.ac.za> wrote:
>
> Thanks for the suggestions I will try it out and let you know how it goes
>
> Kry Outlook vir iOS<https://aka.ms/o0ukef>
> ________________________________
> From: Bill Ross <ross.cgl.ucsf.edu>
> Sent: Monday, September 11, 2017 5:32:01 PM
> To: amber.ambermd.org
> Subject: Re: [AMBER] Minimization error due to modified residue
>
> It would be interesting to try on a GPU if the CPU was facing LINMIN, too.
>
> Bill
>
>
> On 9/11/17 8:09 AM, Shafaq Muzammal wrote:
>> It might be a childish solution but i faced the same problem for a few simulations and was suggested to ignore this error. But running minimisation on CPU solved the issue for me.
>>
>>
>> Best
>>
>>
>> Shafaq
>>
>> ________________________________
>> From: Lizelle Lubbe <LBBLIZ002.myuct.ac.za>
>> Sent: Monday, September 11, 2017 3:59:38 PM
>> To: AMBER Mailing List
>> Subject: Re: [AMBER] Minimization error due to modified residue
>>
>> Okay I will read up more before doing the equilibration. Thanks so much for all the help.
>>
>> Yes Shafaq I am using amber on gpu for production but not minimization. It is run on a cluster
>>
>> Kry Outlook vir iOS<https://aka.ms/o0ukef>
>> ________________________________
>> From: Shafaq Muzammal <Shafaq.Muzammal.nottingham.ac.uk>
>> Sent: Monday, September 11, 2017 4:57:48 PM
>> To: AMBER Mailing List
>> Subject: Re: [AMBER] Minimization error due to modified residue
>>
>> Dear Lizelle,
>>
>>
>> Are you using Amber on a GPU code?
>>
>>
>> Regards,
>>
>> Shafaq
>>
>> ________________________________
>> From: Lizelle Lubbe <LBBLIZ002.myuct.ac.za>
>> Sent: Monday, September 11, 2017 3:48:26 PM
>> To: AMBER Mailing List
>> Subject: Re: [AMBER] Minimization error due to modified residue
>>
>> So are you saying that as long as the min2 structure looks fine although it gives a LINMIN error you can proceed to heating and equilibration?
>>
>> Would you consider this heating input to be slow enough?
>>
>> Heating
>>
>> # Control section
>> &cntrl
>> imin = 0,
>> irest = 0,
>> ntx = 1,
>> ntb = 1, ntp = 0,
>> cut = 10.0,
>> ntc = 2, ntf = 2,
>> tempi = 0.0, temp0 = 300.0,
>> ntt = 3, gamma_ln = 2, ig=-1,
>> nstlim = 50000, dt = 0.002,
>> ntwx = 5000, ntpr = 5000, ntwe = 5000,
>> ioutfm = 1,
>> iwrap = 1,
>> ntr = 1, restraintmask=':1-721,756', restraint_wt=10.0,
>> /
>>
>>
>> Lizelle Lubbe
>>
>> PhD (Medical biochemistry) candidate
>> Department of Integrative Biomedical Sciences
>> University of Cape Town
>> ________________________________________
>> From: Bill Ross <ross.cgl.ucsf.edu>
>> Sent: 11 September 2017 04:37:24 PM
>> To: amber.ambermd.org
>> Subject: Re: [AMBER] Minimization error due to modified residue
>>
>> My mistake. I suspect the LINMIN failure you are seeing now is because
>> of the charge of the Zn, and may be 'just life'. I would try to warm up
>> slowly, and watch the Zn area with VMD.
>>
>> Bill
>>
>>
>> On 9/11/17 7:30 AM, Lizelle Lubbe wrote:
>>> I'm not sure which H you're referring to... Previously the H from WAR was attracted to the OE1 of GLR
>>>
>>> ________________________________________
>>> From: Bill Ross <ross.cgl.ucsf.edu>
>>> Sent: 11 September 2017 04:28:24 PM
>>> To: amber.ambermd.org
>>> Subject: Re: [AMBER] Minimization error due to modified residue
>>>
>>> One thing I didn't think of is, why would an H be attracted to a Zn
>>> anyway, assuming both have positive charge.
>>>
>>> Bill
>>>
>>>
>>> On 9/11/17 7:03 AM, Lizelle Lubbe wrote:
>>>> Okay I just tried that. When trying to minimize the solvent while having ntc=ntf=2 it failed due to LINMIN error
>>>> ___________________________________________________________________________________________________________________
>>>> min1.out:
>>>>
>>>> ... RESTARTED DUE TO LINMIN FAILURE ...
>>>>
>>>>
>>>> NSTEP ENERGY RMS GMAX NAME NUMBER
>>>> 600 -5.5571E+05 1.4422E+01 1.0911E+02 Zn 12493
>>>>
>>>> BOND = 524.4219 ANGLE = 1593.8799 DIHED = 6681.1564
>>>> VDWAALS = 36245.1401 EEL = -651519.6869 HBOND = 0.0000
>>>> 1-4 VDW = 3068.4211 1-4 EEL = 47143.0466 RESTRAINT = 557.1118
>>>> EAMBER = -556263.6209
>>>>
>>>> ... RESTARTED DUE TO LINMIN FAILURE ...
>>>>
>>>> ***** REPEATED LINMIN FAILURE *****
>>>>
>>>> ***** SEE http://ambermd.org/Questions/linmin.html FOR MORE INFO *****
>>>>
>>>>
>>>> FINAL RESULTS
>>>>
>>>>
>>>>
>>>> NSTEP ENERGY RMS GMAX NAME NUMBER
>>>> 614 -5.5571E+05 1.4422E+01 1.0911E+02 Zn 12493
>>>>
>>>> BOND = 524.4219 ANGLE = 1593.8799 DIHED = 6681.1564
>>>> VDWAALS = 36245.1392 EEL = -651519.6849 HBOND = 0.0000
>>>> 1-4 VDW = 3068.4211 1-4 EEL = 47143.0466 RESTRAINT = 557.1117
>>>> EAMBER = -556263.6199
>>>> _______________________________________________________________________________________________________________________________
>>>> This link mentioned that having SHAKE on can cause this error and that initial minimization should be done without it.
>>>>
>>>> When using min1.rst7 from successful minimization1 without SHAKE then minimization 2 with ntc=ntf=2 also failed due to LINMIN error. Atom 5686 is the NE2 of HIR that coordinates the ZN
>>>> Any other ideas as to what I can do to solve this problem?
>>>>
>>>> min2.out:
>>>> NSTEP ENERGY RMS GMAX NAME NUMBER
>>>> 5000 -6.8624E+05 1.9238E+01 1.1379E+02 NE2 5686
>>>>
>>>> BOND = 570.2513 ANGLE = 1577.8284 DIHED = 6671.0734
>>>> VDWAALS = 134678.7301 EEL = -879884.8564 HBOND = 0.0000
>>>> 1-4 VDW = 3049.4894 1-4 EEL = 47099.7304 RESTRAINT = 0.0000
>>>>
>>>> ... RESTARTED DUE TO LINMIN FAILURE ...
>>>>
>>>> ... RESTARTED DUE TO LINMIN FAILURE ...
>>>>
>>>>
>>>> NSTEP ENERGY RMS GMAX NAME NUMBER
>>>> 5050 -6.8624E+05 1.9238E+01 1.1379E+02 NE2 5686
>>>>
>>>> BOND = 570.2511 ANGLE = 1577.8284 DIHED = 6671.0733
>>>> VDWAALS = 134678.7224 EEL = -879884.8458 HBOND = 0.0000
>>>> 1-4 VDW = 3049.4894 1-4 EEL = 47099.7303 RESTRAINT = 0.0000
>>>>
>>>> ... RESTARTED DUE TO LINMIN FAILURE ...
>>>>
>>>> ... RESTARTED DUE TO LINMIN FAILURE ...
>>>>
>>>>
>>>> NSTEP ENERGY RMS GMAX NAME NUMBER
>>>> 5100 -6.8624E+05 1.9238E+01 1.1378E+02 NE2 5686
>>>>
>>>> BOND = 570.2508 ANGLE = 1577.8285 DIHED = 6671.0733
>>>> VDWAALS = 134678.7124 EEL = -879884.8334 HBOND = 0.0000
>>>> 1-4 VDW = 3049.4894 1-4 EEL = 47099.7300 RESTRAINT = 0.0000
>>>>
>>>> ... RESTARTED DUE TO LINMIN FAILURE ...
>>>>
>>>> ***** REPEATED LINMIN FAILURE *****
>>>>
>>>> ***** SEE http://ambermd.org/Questions/linmin.html FOR MORE INFO *****
>>>>
>>>>
>>>> FINAL RESULTS
>>>>
>>>>
>>>>
>>>> NSTEP ENERGY RMS GMAX NAME NUMBER
>>>> 5114 -6.8624E+05 1.9238E+01 1.1378E+02 NE2 5686
>>>>
>>>> BOND = 570.2507 ANGLE = 1577.8285 DIHED = 6671.0733
>>>> VDWAALS = 134678.7100 EEL = -879884.8302 HBOND = 0.0000
>>>> 1-4 VDW = 3049.4894 1-4 EEL = 47099.7299 RESTRAINT = 0.0000
>>>>
>>>>
>>>> Lizelle Lubbe
>>>>
>>>> PhD (Medical biochemistry) candidate
>>>> Department of Integrative Biomedical Sciences
>>>> University of Cape Town
>>>> ________________________________________
>>>> From: Lizelle Lubbe <LBBLIZ002.myuct.ac.za>
>>>> Sent: 11 September 2017 03:12 PM
>>>> To: AMBER Mailing List
>>>> Subject: Re: [AMBER] Minimization error due to modified residue
>>>>
>>>> No I did not. Thought that it should be off for the minimization and then on for equilibration.
>>>> Will try minimization1 now with ntc=2 and ntf=2
>>>>
>>>> Thanks Bill
>>>>
>>>> Kind regards
>>>>
>>>> Lizelle
>>>> ________________________________________
>>>> From: Bill Ross <ross.cgl.ucsf.edu>
>>>> Sent: 11 September 2017 03:09:40 PM
>>>> To: amber.ambermd.org
>>>> Subject: Re: [AMBER] Minimization error due to modified residue
>>>>
>>>> Are you using ntc=2, ntf=2?
>>>>
>>>> This would be required to keep TIP3 H's within the boundary of the O vdw
>>>> radius, since the H's have no radius.
>>>>
>>>> Bill
>>>>
>>>>
>>>> On 9/11/17 5:38 AM, Lizelle Lubbe wrote:
>>>>> the atoms 6143 OE1 of GLR389 (the ZN-coordinated Glu) and 12495 H1 of WAR (the ZN-coordinated TIP3p water) overlap
>>>> _______________________________________________
>>>> AMBER mailing list
>>>> AMBER.ambermd.org
>>>> http://lists.ambermd.org/mailman/listinfo/amber
>>>> Disclaimer - University of Cape Town This e-mail is subject to UCT policies and e-mail disclaimer published on our website at http://www.uct.ac.za/about/policies/emaildisclaimer/ or obtainable from +27 21 650 9111. If this e-mail is not related to the business of UCT, it is sent by the sender in an individual capacity. Please report security incidents or abuse via csirt.uct.ac.za
>>>>
>>>> _______________________________________________
>>>> AMBER mailing list
>>>> AMBER.ambermd.org
>>>> http://lists.ambermd.org/mailman/listinfo/amber
>>>> Disclaimer - University of Cape Town This e-mail is subject to UCT policies and e-mail disclaimer published on our website at http://www.uct.ac.za/about/policies/emaildisclaimer/ or obtainable from +27 21 650 9111. If this e-mail is not related to the business of UCT, it is sent by the sender in an individual capacity. Please report security incidents or abuse via csirt.uct.ac.za
>>>>
>>>> _______________________________________________
>>>> AMBER mailing list
>>>> AMBER.ambermd.org
>>>> http://lists.ambermd.org/mailman/listinfo/amber
>>> _______________________________________________
>>> AMBER mailing list
>>> AMBER.ambermd.org
>>> http://lists.ambermd.org/mailman/listinfo/amber
>>> Disclaimer - University of Cape Town This e-mail is subject to UCT policies and e-mail disclaimer published on our website at http://www.uct.ac.za/about/policies/emaildisclaimer/ or obtainable from +27 21 650 9111. If this e-mail is not related to the business of UCT, it is sent by the sender in an individual capacity. Please report security incidents or abuse via csirt.uct.ac.za
>>>
>>> _______________________________________________
>>> AMBER mailing list
>>> AMBER.ambermd.org
>>> http://lists.ambermd.org/mailman/listinfo/amber
>>
>> _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber
>> Disclaimer - University of Cape Town This e-mail is subject to UCT policies and e-mail disclaimer published on our website at http://www.uct.ac.za/about/policies/emaildisclaimer/ or obtainable from +27 21 650 9111. If this e-mail is not related to the business of UCT, it is sent by the sender in an individual capacity. Please report security incidents or abuse via csirt.uct.ac.za
>>
>> _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber
>>
>>
>>
>>
>> This message and any attachment are intended solely for the addressee
>> and may contain confidential information. If you have received this
>> message in error, please send it back to me, and immediately delete it.
>>
>> Please do not use, copy or disclose the information contained in this
>> message or in any attachment. Any views or opinions expressed by the
>> author of this email do not necessarily reflect the views of the
>> University of Nottingham.
>>
>> This message has been checked for viruses but the contents of an
>> attachment may still contain software viruses which could damage your
>> computer system, you are advised to perform your own checks. Email
>> communications with the University of Nottingham may be monitored as
>> permitted by UK legislation.
>>
>> _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber
>> Disclaimer - University of Cape Town This e-mail is subject to UCT policies and e-mail disclaimer published on our website at http://www.uct.ac.za/about/policies/emaildisclaimer/ or obtainable from +27 21 650 9111. If this e-mail is not related to the business of UCT, it is sent by the sender in an individual capacity. Please report security incidents or abuse via csirt.uct.ac.za
>> _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber
>>
>>
>>
>>
>> This message and any attachment are intended solely for the addressee
>> and may contain confidential information. If you have received this
>> message in error, please send it back to me, and immediately delete it.
>>
>> Please do not use, copy or disclose the information contained in this
>> message or in any attachment. Any views or opinions expressed by the
>> author of this email do not necessarily reflect the views of the
>> University of Nottingham.
>>
>> This message has been checked for viruses but the contents of an
>> attachment may still contain software viruses which could damage your
>> computer system, you are advised to perform your own checks. Email
>> communications with the University of Nottingham may be monitored as
>> permitted by UK legislation.
>>
>> _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
> Disclaimer - University of Cape Town This e-mail is subject to UCT policies and e-mail disclaimer published on our website at http://www.uct.ac.za/about/policies/emaildisclaimer/ or obtainable from +27 21 650 9111. If this e-mail is not related to the business of UCT, it is sent by the sender in an individual capacity. Please report security incidents or abuse via csirt.uct.ac.za
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Sep 11 2017 - 10:30:02 PDT
