[gmx-users] REMD exchange probabilities

Neha Gandhi n.gandhiau at gmail.com
Sun Mar 8 19:25:23 CET 2015


Dear list,

Using an exchange probability of 0.25 and temperature range 293-370  K, I
calculated number of replicas using the server. However, when I did first
run and tried exchanging replicas every 500 steps (1 ps), I don't think the
exchange probabilities make sense in particular replicas 15 and 16. Replica
15 has a low exchange ratio of 0.12 while replica 16 has a high exchange
ratio of 0.55.

Repl  average probabilities:
Repl     0    1    2    3    4    5    6    7    8    9   10   11   12
13   14   15   16   17   18   19   20   21   22   23   24   25   26   27
28   29   30   31   32   33   34   35   36   37   38   39   40   41   42
43   44   45   46   47
Repl      .28  .28  .28  .28  .29  .28  .29  .29  .28  .29  .28  .28  .29
.29  .29  .12  .55  .29  .29  .30  .30  .29  .29  .26  .32  .31  .30  .30
.30  .30  .30  .31  .31  .31  .31  .31  .31  .31  .31  .31  .31  .31  .32
.32  .32  .32  .33
Repl  number of exchanges:
Repl     0    1    2    3    4    5    6    7    8    9   10   11   12
13   14   15   16   17   18   19   20   21   22   23   24   25   26   27
28   29   30   31   32   33   34   35   36   37   38   39   40   41   42
43   44   45   46   47
Repl     2901 2954 2873 3017 3038 2910 3009 2993 2934 3002 2981 2999 2927
3038 3059 1229 5757 3056 3100 3136 3054 3053 3109 2743 3333 3166 3097 3185
3161 3189 3133 3226 3261 3242 3229 3205 3249 3227 3221 3222 3326 3303 3309
3320 3373 3346 3474
Repl  average number of exchanges:
Repl     0    1    2    3    4    5    6    7    8    9   10   11   12
13   14   15   16   17   18   19   20   21   22   23   24   25   26   27
28   29   30   31   32   33   34   35   36   37   38   39   40   41   42
43   44   45   46   47
Repl      .28  .28  .27  .29  .29  .28  .29  .29  .28  .29  .29  .29  .28
.29  .29  .12  .55  .29  .30  .30  .29  .29  .30  .26  .32  .30  .30  .30
.30  .31  .30  .31  .31  .31  .31  .31  .31  .31  .31  .31  .32  .32  .32
.32  .32  .32  .33


Below are the temperatures I have used. How do I manually edit temperatures
to get average exchange probabilities between 0.2-0.3?

ref_t        = 293            293    ; reference temperature, one for each
group, in K
ref_t        = 294.51     294.51    ; reference temperature, one for each
group, in K
ref_t        = 296.03     296.03    ; reference temperature, one for each
group, in K
ref_t        = 297.56     297.56    ; reference temperature, one for each
group, in K
ref_t        = 299.09     299.09    ; reference temperature, one for each
group, in K
ref_t        = 300.63     300.63    ; reference temperature, one for each
group, in K
ref_t        = 302.18     302.18    ; reference temperature, one for each
group, in K
ref_t        = 303.73     303.73    ; reference temperature, one for each
group, in K
ref_t        = 305.29     305.29    ; reference temperature, one for each
group, in K
ref_t        = 306.86     306.86    ; reference temperature, one for each
group, in K
ref_t        = 308.43     308.43    ; reference temperature, one for each
group, in K
ref_t        = 310.01     310.01    ; reference temperature, one for each
group, in K
ref_t        = 311.60     311.60    ; reference temperature, one for each
group, in K
ref_t        = 313.19     313.19    ; reference temperature, one for each
group, in K
ref_t        = 314.79     314.79    ; reference temperature, one for each
group, in K
ref_t        = 316.40     316.40    ; reference temperature, one for each
group, in K
ref_t        = 318.63     318.63    ; reference temperature, one for each
group, in K
ref_t        = 319.63     319.63    ; reference temperature, one for each
group, in K
ref_t        = 321.26     321.26    ; reference temperature, one for each
group, in K
ref_t        = 322.89     322.89    ; reference temperature, one for each
group, in K
ref_t        = 324.52     324.52    ; reference temperature, one for each
group, in K
ref_t        = 326.17     326.17    ; reference temperature, one for each
group, in K
ref_t        = 327.82     327.82    ; reference temperature, one for each
group, in K
ref_t        = 329.49     329.49    ; reference temperature, one for each
group, in K
ref_t        = 331.26     331.26    ; reference temperature, one for each
group, in K
ref_t        = 332.86     332.86    ; reference temperature, one for each
group, in K
ref_t        = 334.51     334.51    ; reference temperature, one for each
group, in K
ref_t        = 336.20     336.20    ; reference temperature, one for each
group, in K
ref_t        = 337.90     337.90    ; reference temperature, one for each
group, in K
ref_t        = 339.61     339.61    ; reference temperature, one for each
group, in K
ref_t        = 341.32     341.32    ; reference temperature, one for each
group, in K
ref_t        = 343.04     343.04    ; reference temperature, one for each
group, in K
ref_t        = 344.76     344.76    ; reference temperature, one for each
group, in K
ref_t        = 346.49     346.49    ; reference temperature, one for each
group, in K
ref_t        = 348.23     348.23    ; reference temperature, one for each
group, in K
ref_t        = 349.98     349.98    ; reference temperature, one for each
group, in K
ref_t        = 351.74     351.74    ; reference temperature, one for each
group, in K
ref_t        = 353.50     353.50    ; reference temperature, one for each
group, in K
ref_t        = 355.27     355.27    ; reference temperature, one for each
group, in K
ref_t        = 357.04     357.04    ; reference temperature, one for each
group, in K
ref_t        = 358.83     358.83    ; reference temperature, one for each
group, in K
ref_t        = 360.62     360.62    ; reference temperature, one for each
group, in K
ref_t        = 362.42     362.42    ; reference temperature, one for each
group, in K
ref_t        = 364.22     364.22    ; reference temperature, one for each
group, in K
ref_t        = 366.04     366.04    ; reference temperature, one for each
group, in K
ref_t        = 367.86     367.86    ; reference temperature, one for each
group, in K
ref_t        = 369.69     369.69    ; reference temperature, one for each
group, in K
ref_t        = 371.52     371.52    ; reference temperature, one for each
group, in K

P.S. I have realised that the longer the time spent in each replica the
better is the sampling. I should exchange replica every 1250 steps x
exchange probability of 0.25 to get 10 ps of time spent in every replica.


-- 
Regards,
Dr. Neha S. Gandhi,
Curtin Research Fellow,
School of Biomedical Sciences,
Curtin University,
Perth GPO U1987
Australia
LinkedIn
Research Gate


More information about the gromacs.org_gmx-users mailing list