超胞和溶液的共同盒子发生能量极小化异常和动力学报错

syarnold

各位老师，我使用packmol对超胞和溶液生成了一个盒子，然后通过GROMACS跑动力学模拟的过程中遇到了报错，其中能量极小化异常如下：

WARNING: Listed nonbonded interaction between particles 8099 and 8101
at distance 2.884 which is larger than the table limit 1.945 nm.


This is likely either a 1,4 interaction, or a listed interaction inside
a smaller molecule you are decoupling during a free energy calculation.
Since interactions at distances beyond the table cannot be computed,
they are skipped until they are inside the table limit again. You will
only see this message once, even if it occurs for several interactions.


IMPORTANT: This should not happen in a stable simulation, so there is
probably something wrong with your system. Only change the table-extension
distance in the mdp file if you are really sure that is the reason.




Step 20, Epot=2.800657e+07, Fnorm=8.503e+05, Fmax=1.485e+07 (atom 6581)
Step 21, Epot=4.157206e+10, Fnorm=3.610e+10, Fmax=1.525e+12 (atom 3475)
Step 22, Epot=1.044017e+13, Fnorm=2.261e+13, Fmax=1.575e+15 (atom 3308)
Step 23, Epot=1.194615e+12, Fnorm=2.991e+12, Fmax=2.292e+14 (atom 5029)
Step 24, Epot=6.779990e+12, Fnorm=1.838e+13, Fmax=1.546e+15 (atom 2676)
Step 25, Epot=2.830741e+16, Fnorm=      inf, Fmax=      inf (atom 2479) 有时候会在这步卡主，有时候会过去，过去后会提示如下：


Energy minimization has stopped, but the forces have not converged to the
requested precision Fmax < 100 (which may not be possible for your system).
It stopped because the algorithm tried to make a new step whose size was too
small, or there was no change in the energy since last step. Either way, we
regard the minimization as converged to within the available machine
precision, given your starting configuration and EM parameters.


Double precision normally gives you higher accuracy, but this is often not
needed for preparing to run molecular dynamics.


writing lowest energy coordinates.


Polak-Ribiere Conjugate Gradients converged to machine precision in 129 steps,
but did not reach the requested Fmax < 100.
Potential Energy  =  3.4181414e+08
Maximum force     =  2.2467842e+08 on atom 12096
Norm of force     =  2.3482725e+06


跑动力学则发生过两种报错：
第一种：
Program:     gmx mdrun, version 2020.6-MODIFIED
Source file: src\gromacs\pbcutil\mshift.cpp (line 904)

Fatal error:
There are inconsistent shifts over periodic boundaries in a molecule type
consisting of 304 atoms. The longest distance involved in such interactions is
2.286 nm which is above half the box length. Either you have excessively large
distances between atoms in bonded interactions or your system is exploding.


第二种：
Fatal error: Step 0: The total potential energy is 1.24207e+14, which is extremely high.

我在之前和之后都试过将完全相同的超胞和溶液盒子分别进行动力学模拟，都可以正常进行计算，但是合起来就不行，还请各位大佬指点迷津。

sobereva

此类问题照着http://sobereva.com/soft/Sobtop#FAQ8一点点排查

syarnold

sobereva 发表于 2025-2-21 07:35
此类问题照着http://sobereva.com/soft/Sobtop#FAQ8一点点排查

感谢社长的指点，经过不断的排查，最终发现并解决了问题。

问题还是出在结构上，这里帮同为新手的小伙伴避个坑，结构的问题出在超胞上。晶胞本身是没有问题，但是由于扩胞过程中边界会截断一些分子，导致一些游离原子“漂”在边界附近。如果是单独超胞的盒子（周期性边界紧挨着超胞）那是可以直接跑MD的，但如果像我这样在超胞的某个边界另加了其他分子（扩大了周期性边界，不在超胞边上了），那么将可能导致体系出现崩溃。

解决方法是先跑一个单独超胞盒子的MD，然后从结果中选一帧所有分子都完整的另存下来，再以此为基础重新生成新的超胞+溶剂的盒子。