ERROR: the system is looking for a value outside the grid along the 0 (dk) index!

122 views
Skip to first unread message

Christopher Joo

unread,
Nov 15, 2023, 8:52:59 AM11/15/23
to PLUMED users
Hello plumed users,

I got an error when using metadynamics with PLUMED2.5.
ERROR: the system is looking for a value outside the grid along the 0 (dk) index!

It shows that the CV is out of the grid max. But I have increased the grid max value to the twice of CV limit restraint by UPPER WALLS. The simulation is still crashed after running for some time. I have checked the CV value right after the simulation stops. It is still smaller than the limit set in UPPER WALLS. 

Here is my plumed script, 

RESTART
UNITS LENGTH=A TIME=fs ENERGY=kj/mol

pca: COM ATOMS=442
#pkik: COM ATOMS=1026,1029,124,134,137
#pkik: COM ATOMS=127,444,1331,1015,1336
kiknCw: COM ATOMS=127,444,1331,1015,1336
pkinKCw: POSITION ATOM=kiknCw
ppCa: POSITION ATOM=442
dx: CUSTOM ARG=ppCa.x,pkinKCw.x FUNC=x-(y-0.0125969552238805) PERIODIC=NO
dy: CUSTOM ARG=ppCa.y,pkinKCw.y FUNC=x-(y-0.0258486865671641) PERIODIC=NO
dz: CUSTOM ARG=ppCa.z,pkinKCw.z FUNC=x-(y+4.90699713432835) PERIODIC=NO
 VAR=ax,ay,az,bx,by,bz FUNC=sqrt((ax-(bx-0.003832961))*(ax-(bx-0.003832961))+(ay-(by+2.374226467))*(ay-(by+2.374226467))+(az-(bz+0.010088556))*(az-(bz+0.010088556))) PERIODIC=NO
dk: CUSTOM ARG=dx,dy,dz FUNC=sqrt(x*x+y*y+z*z) PERIODIC=NO

Ca: GROUP ATOMS=442
Cw: GROUP ATOMS=124,128,1016
poCw1: POSITION ATOM=124
poCw2: POSITION ATOM=128
poCw3: POSITION ATOM=1016

Ow: GROUP ATOMS=1864-2913:3
Hw: GROUP ATOMS=1864-2913 REMOVE=Ow
C: GROUP ATOMS=1861
Oc: GROUP ATOMS=1862,1863
O: GROUP ATOMS=1864-2913:3,106,108,110,112,146,148,150,152,426,428,430,432,439,455,456,457,458,459,461,463,723,725,727,729,759,760,761,763,764,767,768,1036,1038,1040,1042,1043,1047,1076,1078,1080,1082,1345,1346,1347,1348,1349,1350,1351,1352,1378,1380,1382,1384,1648,1650,1651,1653,1656,1660,1687,1689,1691,1693,114,116,118,120,154,156,158,160,239,434,436,438,440,460,462,524,530,532,534,536,731,733,735,737,766,770,837,839,841,845,847,1044,1046,1048,1050,1084,1086,1088,1090,1117,1123,1127,1386,1388,1390,1392,1414,1416,1418,1420,1422,1424,1450,1655,1657,1659,1661,1695,1697,1699,1701,1725,1729
H: GROUP ATOMS=1864-2913,90,92,94,96,130,132,134,136,410,412,414,416,423,446,447,448,449,450,452,454,707,709,711,713,746,747,748,750,751,754,755,1020,1022,1024,1026,1027,1031,1060,1062,1064,1066,1337,1338,1339,1340,1341,1342,1343,1344,1362,1364,1366,1368,1634,1636,1637,1639,1642,1646,1671,1673,1675,1677,98,100,102,104,138,140,142,144,223,418,420,422,424,451,453,510,516,518,520,522,715,717,719,721,753,757,821,823,825,829,831,1028,1030,1032,1034,1068,1070,1072,1074,1101,1107,1111,1370,1372,1374,1376,1400,1402,1404,1406,1408,1410,1434,1641,1643,1645,1647,1679,1681,1683,1685,1709,1713 REMOVE=Ow

cn: COORDINATION GROUPA=Ca GROUPB=Cw SWITCH={RATIONAL NN=8 MM=16 D_0=0 R_0=4.5}
cnCaOw: COORDINATION GROUPA=Ca GROUPB=O SWITCH={RATIONAL NN=8 MM=16 D_0=0 R_0=2.8}

lwall: LOWER_WALLS ARG=dy,dz AT=-2,-2 KAPPA=200.0,200.0 EXP=2,2 EPS=1,1 OFFSET=0,0

ucnwall: UPPER_WALLS ARG=dk,cn AT=6,3.0 KAPPA=200.0,200.0 EXP=2,2 EPS=1,1 OFFSET=0,0

METAD ...
LABEL=metad
ARG=dk,cn SIGMA=0.1,0.1 HEIGHT=1.2 PACE=500 TEMP=300.0 BIASFACTOR=20
GRID_MIN=0,0 GRID_MAX=12,4 GRID_BIN=600,200
CALC_RCT
FILE=HILLS
... METAD

PRINT STRIDE=500 ARG=dx,dy,dz,dk,cn,cnCaOw,lwall.bias,ucnwall.bias,metad.bias,metad.rbias,metad.rct  FILE=COLVAR

How to solve this issue? 

Thanks in advance.

Best,

Joo

Giovanni Bussi

unread,
Nov 15, 2023, 9:21:35 AM11/15/23
to plumed...@googlegroups.com
It's normal that the value *before* the crash is lower than the boundary. Indeed, as soon as the value becomes higher than the boundary the simulation will stop, and the COLVAR file will be lagging behind to to the STRIDE used for printing it or to filesystem cashing.

If this is a distance, you should estimate the maximum value based on the cell size.

Giovanni


--
You received this message because you are subscribed to the Google Groups "PLUMED users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to plumed-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/plumed-users/b84eff86-6873-4980-b87e-66f6657a4973n%40googlegroups.com.
Reply all
Reply to author
Forward
0 new messages