计算化学公社

标题: XTB出新版咯：GFN2-xTB，大家踊跃申请使用哦 [打印本页]

作者
Author: yjmaxpayne 时间: 2018-3-31 18:18
标题: XTB出新版咯：GFN2-xTB，大家踊跃申请使用哦
我只是邮件搬运工。。。

The main new features are
- multipole electrostatics
- removal of all classical corrections as well as special pair
potentials/parameters,
. i.e., its less empirical
- ATM dispersion term
- atom specific CN dependent level shift and one-center XC-anisotropy term
- much better parameterization (more reference systems, better frequency
fit)
What is still missing (but likely will have only a small effect at this
accuracy level)
is the D4 instead of D3 dispersion model which will be implemented
soon. We are also working on the GFN2 paper but this may take some weeks.

Many properties and problems have been improved or solved compared
to GFN1. Most notably are non-covalent interactions and conformational
energies improved
and most metals are described more consistently. However, there are
of course also drawbacks and on the downside we note sometimes less
stable SCF
convergence behavior. Both, this and the higher accuracy is related to
the short range part of the
multipole electrostatics. In non-convergent-SCF cases GFN1 is still
available as fall-back option.

The new method is on average only insignificantly slower than GFN1 (even
faster for systems with many hydrogens) and we think that GFN2-xTB
offers an extremely good cost/performance ratio often reaching
DFT-D3 accuracy for the target properties. All non-covalent interactions
are now exceptionally good including water, and halogen or pnicogen
interactions.
Additionally some typical thermochemistry properties are described better
(without fitting them!) which will be discussed in detail in the
forthcoming paper.
In summary I don't see much room for further general improvement at this
high-speed QM level.

To run GFN2 just copy the parameter file .param_gfn2.xtb
to the xtb path and set the default in the .xtbrc file
(line two). GFN1 is still available by the flag -gfn1
(or setting the default of variable gfnver to 1).

Important note: the parameters are preliminary, i.e.,
it make little sense to start big projects @GFN2 until
we got feedback and may be change a few details
(which is more likely to happen for heavier, "uncommon" elements
than for HBCNOFCl.. for which the present parameters
are probably the final ones). In addition we're right now
re-adjusting the GBSA parameters to GFN2. The present GFN1
GBSA parameters can be used, however, without
much loss of accuracy.

Please report any problems, bugs, or bad cases
but we're also thankful for success stories
(Christoph Bannwarth and I invested some time into all this and
if was often but not always fun ...).
Those of you who sent me GFN1 outliers (thanks again!)
may check if they are now qualitatively ok with GFN2 or
at least have improved.

作者
Author: ikea1984 时间: 2018-4-2 22:15
啥时候搞出周期性版本的就流弊了，吊打DFTB啊

作者
Author: sobereva 时间: 2018-4-10 22:24
最近Grimme动作很频繁
Dear xtb users,
below you can find a link to the second preliminary test version
of the code implementing GFN2-xTB, an improved version of the GFN-xTB
method. The changes with respect to the previous test version are
- re-fit of all elements except HCNO (though with mostly small
parameter changes)
- D4 instead of D3 dispersion
This release is supposed to be close to the final version.
To run GFN2 just copy the parameter file .param_gfn2.xtb
to the xtb path and set the default in the .xtbrc file
(line two, already set properly to the latest version).
Please report any problems, bugs, or bad cases to me.
Best
Stefan Grimme

作者
Author: Warm_Cloud 时间: 2018-4-11 08:46
大体系才是王道。。。

作者
Author: yjmaxpayne 时间: 2018-4-11 10:53
最近一直在用GFN2-XTB，simulated annealing找conformer确实给力。但是似乎GBSA水环境优化带水壳的分子系统还是有问题（详细见社长的帖子：http://bbs.keinsci.com/thread-9129-1-1.html）等我有空写信给Grimme反映一下。

作者
Author: sobereva 时间: 2018-4-27 17:07
Dear xtb users,
below you can find the download link to the latest GFN2-xTB test version.
This version contains new parameters for all elements as well as
a few bug fixes. The D4 dispersion correction is now fully (and
correctly) included and operational. The main change to the
last version is an increased H0 flexibility by l-dependent KCN
parameters and hand-adjusted (smaller) AO exponents for H and
C giving better Pauli repulsion. I hope that this was the
last bigger change and that this version will be the finally
published one. What is still missing is the re-fit of the
GBSA solvation model for GFN2 (although the GBSA
files for GFN1 can actually be used for GFN2 as well).
Best
Stefan Grimme

作者
Author: hakuna 时间: 2018-4-27 19:15
本帖最后由 hakuna 于 2018-4-27 19:17 编辑

刚收到的

Dear xtb users,
below you can find the download link to the latest GFN2-xTB test version.
This version contains new parameters for all elements as well as
a few bug fixes. The D4 dispersion correction is now fully (and
correctly) included and operational. The main change to the
last version is an increased H0 flexibility by l-dependent KCN
parameters and hand-adjusted (smaller) AO exponents for H and
C giving better Pauli repulsion. I hope that this was the
last bigger change and that this version will be the finally
published one. What is still missing is the re-fit of the
GBSA solvation model for GFN2 (although the GBSA
files for GFN1 can actually be used for GFN2 as well).

Best
Stefan Grimme

作者
Author: hzfish 时间: 2018-8-14 07:32
Dear xtb users,
during the last months we worked on the GFN2 paper (appearing soon on Chemarxiv),
removed some very small bugs and slightly revised (triggered
by user input) the GFN2 parameters for Se and I.
See file REVISION for details.
We also had to revise the scripts and programs for the automated NMR spectra
calculation. This will be finished soon and announced through this user list.
To download the current version just use the link below.
Best
Stefan Grimme

作者
Author: Accelerator 时间: 2018-12-6 00:57
最近尝试用GFN2-xTB跑分子动力学（用DFT很难算动），发现虽然在特定结构上的单点能精度不错，跑出来的轨迹行为还是和DFT相差不少的。此外GFN2-xTB会破坏甲醇的氢键网络（甲醇分子构成的团簇全都跑散掉了），类似的还会显著低估DMSO团簇的密度，总之感觉并没有想象的那么美好。

作者
Author: yjmaxpayne 时间: 2018-12-6 17:18

Accelerator 发表于 2018-12-6 00:57
最近尝试用GFN2-xTB跑分子动力学（用DFT很难算动），发现虽然在特定结构上的单点能精度不错，跑出来的轨迹 ...

估计训练集不够吧，另外之前Sob社长提到的问题，即gbsa水环境下优化水团簇的问题，也没完全解决。

作者
Author: IIIIO_OIIII 时间: 2019-2-12 09:26
本帖最后由 IIIIO_OIIII 于 2019-2-12 09:45 编辑

yjmaxpayne 发表于 2018-4-11 10:53
最近一直在用GFN2-XTB，simulated annealing找conformer确实给力。但是似乎GBSA水环境优化带水壳的分子系 ...

楼主好。我最近刚开始使用xtb, 还不是很熟悉怎么使用。现在只会一些比较简单的，比如优化或者社长在博文里演示过的。看到您说可以做simulated annealing 来实现conformer search，所以斗胆问您一句，可以让我看一下您那一部分的命令吗？是通过在xyz的文件末尾加入一些命令（$set … siman … $end)实现的，是吗？但是我看新版的manual里，可选的的命令很少，所以不太清楚该如何组织。我用那有限的命令运行了几次，总是一秒钟就normal termination，有点懵。我对程序的语言不是很会，还请多多指教。或者您给我指一条路，我自己去看也好。不胜感激！

作者
Author: Novice 时间: 2019-3-13 14:28

IIIIO_OIIII 发表于 2019-2-12 09:26
楼主好。我最近刚开始使用xtb, 还不是很熟悉怎么使用。现在只会一些比较简单的，比如优化或者社长在博文 ...

我也是一个XTB新手，看手册感觉无从下手。请问怎么用XTB做simulated annealing 来实现conformer search您弄懂了吗，我也想学习一下，还有我收到邮件有如下说法
“The new conformer search procedure is implemented fully OMP parallel in a code dubbed CREST (Conformer-Rotamer Ensemble Sampling Tool) which can be downloaded as well. The old MF-MD-GC version is still available. ”
这个新的CREST您了解吗？

作者
Author: IIIIO_OIIII 时间: 2019-3-13 15:20

Novice 发表于 2019-3-13 14:28
我也是一个XTB新手，看手册感觉无从下手。请问怎么用XTB做simulated annealing 来实现conformer search您 ...

最近在忙别的，也没顾上搞这个。所以我还是当时的状态，不懂。帮不到你了。估计你要自己先尝试着弄。要是有同学能商量着一起作，估计会好作一些。

作者
Author: Novice 时间: 2019-3-13 15:21

IIIIO_OIIII 发表于 2019-3-13 15:20
最近在忙别的，也没顾上搞这个。所以我还是当时的状态，不懂。帮不到你了。估计你要自己先尝试着弄。要是 ...

哦哦，多谢多谢

作者
Author: IIIIO_OIIII 时间: 2019-3-13 15:42

Novice 发表于 2019-3-13 15:21
哦哦，多谢多谢

如果我有什么进展，我会回复给你。你为什么不用Molclus作conformer search呀？体系太大了吗？

作者
Author: Novice 时间: 2019-3-13 15:51

IIIIO_OIIII 发表于 2019-3-13 15:42
如果我有什么进展，我会回复给你。你为什么不用Molclus作conformer search呀？体系太大了吗？

我经常做的是单分子的构象搜索，以前都是用Win下的GMMX进行的，感觉分子稍微大一点就太慢了。我看社长教程里构象搜索退火用的是Amber，最近新装工作站，装软件弄得我心力憔悴，就不想再折腾了，所以想试试在工作站上用装好的XTB做退火，然后结合molclus做构象搜索筛选

作者
Author: sobereva 时间: 2019-3-14 04:50

Novice 发表于 2019-3-13 15:51
我经常做的是单分子的构象搜索，以前都是用Win下的GMMX进行的，感觉分子稍微大一点就太慢了。我看社长教 ...

"molclus得结合amber才能用"这是天大的误解
正是为了避免使用者非得会MD才能用molclus，后来特意开发了gentor和genmer，作为molclus包中的重要组成部分。有了这俩东西，大多数使用molclus的情况完全不需要牵扯到MD程序
gentor：扫描方式做分子构象搜索的便捷工具
http://bbs.keinsci.com/thread-2388-1-1.html
genmer：生成团簇初始构型的超便捷工具
http://bbs.keinsci.com/thread-2369-1-1.html

molclus一直在不断更新和改进，博文内容也在不断更新和修改

欢迎光临计算化学公社 (http://bbs.keinsci.com/)