对含金属和配体水分子的蛋白-小分子体系进行能量最小化只进行了107步

YuniJ

如题所示，对含有金属锌离子和配体水分子的蛋白-小分子体系进行最小化，只进行了107步，并提示

Energy minimization has stopped, but the forces have not converged to therequested 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.

Steepest Descents converged to machine precision in 108 steps,
but did not reach the requested Fmax < 100.
Potential Energy  = -4.8713222e+05
Maximum force     =  8.4848664e+04 on atom 5098
Norm of force     =  6.1321171e+02

查看了能量最小化的gro构象，发现配体水分子疑似重叠（图1），而中和体系后的gro构象是正常的（图2），根据这个重新修改了配位水分子的拓扑，按照填充溶剂水分子的拓扑参数进行了限制，但是还是出现这样的问题，我上传了top文件和参数文件，请问各位大佬我应该怎么做？

YuniJ

student0618 发表于 2024-6-20 23:58
Past members in my group used the nonbonded model with Li-Merz ion parameters for the active site of ...

非常感谢您！

student0618

Past members in my group used the nonbonded model with Li-Merz ion parameters for the active site of human cyclic nucleotide phosphodiesterase, only hydroxide parameters from literature/resp fitting (not a bacteria phosphotriesterase of your case, but has a similar two-metal site and a bridging hydroxide proposed to be the attacking nucleophile. Similar to the suggested mechanism from the original paper of the PDB you use, but the coordination geometry could differ). At that time this nonbonded model was sufficient for our purpose. However it was before 2012, over 10 years ago, on a different system than yours. I am not sure if this could apply to your system for your purpose of modelling.

Technically mcpb.py can be used to prepare or derive parameters for many metalloenzyme, mainly for the bonded model. Again I am not sure if the bonded model or nonbonded model is better for your system and your purpose, you may refer to some more recent literature of your system to see if anyone has a reasonable way to model it, or test it yourself to see if it gives you something close to crystal structure in MD.

( 抱歉，我不太擅長用中文寫有技術性專有名詞的句子，所以用英文回)

YuniJ

student0618 发表于 2024-6-20 18:40
Not sure if its the reason, but I observed two possible issues in your syetem.

1. from the refer ...

Can I use mcpb.py to replace the water molecule H2O with a hydroxide ion OH- as the bridging ligand between two zinc ions?

YuniJ

student0618 发表于 2024-6-20 18:40
Not sure if its the reason, but I observed two possible issues in your syetem.

1. from the refer ...

thanks！

student0618

Not sure if its the reason, but I observed two possible issues in your syetem.

1. from the reference publication of PDB entry 1HZY (check the Literature box within PDB site or abstract at https://pubs.acs.org/doi/10.1021/bi002661e):

... and a carboxylated Lys 169. Both Lys 169 and a water molecule (or hydroxide ion) serve to bridge the two zinc ions together.

Lys169 (Lys135 in your gro) is carboxylated, but you modelled the lysine and carboxylate separately.

2. A side note here, the oxygen between two metal ions in phosphotriesterase/phosphodiesterase is commonly modelled as hydroxide ion (OH-), maybe also check if you should do this for your complex.