Tanya -
As far as size goes, pmemd will definitely be able to handle your system,
and sander probably would also be okay, but pmemd is more optimized for
large systems. We routinely run systems out around 250,000 atoms. In the
past I have run systems with 500,000 atoms just to make sure we can do it.
In the last release or so, I have been doing things that require more memory
in the code (so for instance I no longer compress the pairlist); I have
headed in this direction because I get better performance and the reality of
large systems is that unless you can run them on substantial numbers of
processors, you have a long wait for results. Now, with large numbers of
processors, the memory requirements go down, so you can run larger systems.
So anyway, I would not worry about memory that much if you are running in
parallel, and if you are not it's execution time you will notice as you head
out around 250K+ atoms.
Regards - Bob Duke
----- Original Message -----
From: "Tanya Johannsen" <tanyajohannsen.yahoo.com>
To: <amber.scripps.edu>
Sent: Wednesday, July 05, 2006 10:29 PM
Subject: AMBER: working with large strained system
> Dear AMBER community:
>
> I was hoping that someone could give me advice about
> AMBER's capabilities for working with a very big, very
> strained system. I have a system of about 23,000 heavy
> atoms (and it will be a lot bigger after I add
> explicit solvent). The system is not a simple crystal
> structure; we have actually performed some refinement
> procedures on it to test a hypothesized conformational
> change, and the resulting structure is rough, with a
> lot of atoms that are close together and things
> twisted in strange ways. I am hoping that an AMBER
> minimization can correct these problems with the
> structure. I have run a test implicit solvent
> minimization on it to see if AMBER can work with the
> system, and since I am new to AMBER and chemistry in
> general, and I am not sure how to interpret the
> results. The initial energy seems high, but the energy
> then decreases in a manner that seems reasonable,
> which makes me wonder if AMBER is managing the system
> okay. I have cut and pasted the energy information
> from the .out file below. If anyone could take a look
> and give me their opinion on whether AMBER can work
> with my structure, I would be very grateful.
>
> Thank you in advance.
>
> Tanya
>
>
> NSTEP ENERGY RMS GMAX
> NAME NUMBER
> 1 1.8815E+11 7.9681E+09 5.0864E+11
> CE1 14332
>
> BOND = 55589.3592 ANGLE = 12184.7419
> DIHED = 24774.6027
> VDWAALS = ************* EEL = -182470.8752 EGB
> = -33528.9928
> 1-4 VDW = 2621189.7868 1-4 EEL = 113699.2076
> RESTRAINT = 0.0000
>
>
> NSTEP ENERGY RMS GMAX
> NAME NUMBER
> 50 8.3833E+04 6.0544E+01 1.1389E+04
> HB3 3770
>
> BOND = 75744.9071 ANGLE = 25839.7402
> DIHED = 26387.7500
> VDWAALS = 41999.7549 EEL = -180985.1851 EGB
> = -35541.1377
> 1-4 VDW = 17095.5974 1-4 EEL = 113291.9949
> RESTRAINT = 0.0000
>
>
> NSTEP ENERGY RMS GMAX
> NAME NUMBER
> 100 -7.6106E+04 6.7496E+00 9.2599E+02
> CG 3771
>
> BOND = 3106.6981 ANGLE = 9954.1219
> DIHED = 25599.3076
> VDWAALS = -16534.4570 EEL = -184943.6352 EGB
> = -35606.2726
> 1-4 VDW = 10899.0169 1-4 EEL = 111419.0978
> RESTRAINT = 0.0000
>
>
> NSTEP ENERGY RMS GMAX
> NAME NUMBER
> 150 -8.4320E+04 1.2693E+00 1.4612E+02
> CZ 20433
>
> BOND = 1835.2873 ANGLE = 7676.0222
> DIHED = 25448.0611
> VDWAALS = -19223.5408 EEL = -183817.5302 EGB
> = -37124.0771
> 1-4 VDW = 9931.3107 1-4 EEL = 110954.2732
> RESTRAINT = 0.0000
>
>
> NSTEP ENERGY RMS GMAX
> NAME NUMBER
> 200 -8.7352E+04 7.4818E-01 5.9198E+01
> CZ 20433
>
> BOND = 1635.7310 ANGLE = 6964.6978
> DIHED = 25387.4683
> VDWAALS = -20400.7798 EEL = -183912.3048 EGB
> = -37147.9082
> 1-4 VDW = 9489.7950 1-4 EEL = 110631.1782
> RESTRAINT = 0.0000
>
>
> NSTEP ENERGY RMS GMAX
> NAME NUMBER
> 250 -8.9041E+04 1.4376E+00 2.5339E+02
> CZ 20433
>
> BOND = 1585.4073 ANGLE = 6589.9043
> DIHED = 25333.5975
> VDWAALS = -21062.1356 EEL = -183477.5794 EGB
> = -37639.9528
> 1-4 VDW = 9244.6705 1-4 EEL = 110385.5562
> RESTRAINT = 0.0000
>
>
> NSTEP ENERGY RMS GMAX
> NAME NUMBER
> 300 -9.0398E+04 2.0106E+00 3.2813E+02
> CZ 20433
>
> BOND = 1579.8403 ANGLE = 6330.0117
> DIHED = 25263.8565
> VDWAALS = -21473.3804 EEL = -183598.8337 EGB
> = -37754.2687
> 1-4 VDW = 9094.2061 1-4 EEL = 110160.1273
> RESTRAINT = 0.0000
>
>
> NSTEP ENERGY RMS GMAX
> NAME NUMBER
> 350 -9.1548E+04 1.3105E+00 2.1914E+02
> CZ 20433
>
> BOND = 1506.5535 ANGLE = 6163.2312
> DIHED = 25173.4086
> VDWAALS = -21679.4755 EEL = -184226.4901 EGB
> = -37502.3111
> 1-4 VDW = 8996.4213 1-4 EEL = 110020.9911
> RESTRAINT = 0.0000
>
>
> NSTEP ENERGY RMS GMAX
> NAME NUMBER
> 400 -9.2264E+04 6.7804E-01 1.1356E+02
> CZ 20433
>
> BOND = 1479.0719 ANGLE = 6046.3070
> DIHED = 25102.9873
> VDWAALS = -21872.8605 EEL = -183766.3511 EGB
> = -38142.6346
> 1-4 VDW = 8911.5434 1-4 EEL = 109978.1407
> RESTRAINT = 0.0000
>
>
> NSTEP ENERGY RMS GMAX
> NAME NUMBER
> 450 -9.2786E+04 1.0530E+00 1.8439E+02
> CZ 20433
>
> BOND = 1477.7497 ANGLE = 5959.9209
> DIHED = 25042.8409
> VDWAALS = -22037.0124 EEL = -182406.5580 EGB
> = -39603.2214
> 1-4 VDW = 8846.2348 1-4 EEL = 109933.6735
> RESTRAINT = 0.0000
>
>
> NSTEP ENERGY RMS GMAX
> NAME NUMBER
> 500 -9.3256E+04 3.2073E-01 1.8957E+01
> CZ 20433
>
> BOND = 1453.1128 ANGLE = 5884.2725
> DIHED = 24989.7304
> VDWAALS = -22165.3787 EEL = -181834.5719 EGB
> = -40269.7803
> 1-4 VDW = 8794.8717 1-4 EEL = 109892.0801
> RESTRAINT = 0.0000
>
>
> Maximum number of minimization cycles reached.
>
>
> FINAL RESULTS
>
>
>
> NSTEP ENERGY RMS GMAX
> NAME NUMBER
> 500 -9.3256E+04 3.2073E-01 1.8957E+01
> CZ 20433
>
> BOND = 1453.1128 ANGLE = 5884.2725
> DIHED = 24989.7304
> VDWAALS = -22165.3787 EEL = -181834.5719 EGB
> = -40269.7803
> 1-4 VDW = 8794.8717 1-4 EEL = 109892.0801
> RESTRAINT = 0.0000
>
>
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam? Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber.scripps.edu
> To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
>
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Sun Jul 09 2006 - 06:07:14 PDT
