计算化学公社

标题: RESP计算无法读取多个构象 [打印本页]

作者
Author: wuzhiyi 时间: 2019-11-12 01:16
标题: RESP计算无法读取多个构象
我想用Multiwfn的RESP功能计算多构象的加权RESP，于是生产甲酸的两个构象，命名为COOH_1.chk和COOH_2.chk。同时生成权重文件conformer.dat

COOH_1.chk 0.5
COOH_2.chk 0.5

随后用Multiwfn生产加权RESP，生产输入文件RESP_2.in

7 // Population analysis and atomic charges
18 // Restrained ElectroStatic Potential (RESP) atomic charge
-1 // Load list of conformer and weights from external file
conformer.dat
1 // Start standard two-stage RESP fitting calculation

执行Multiwfn3.6
Multiwfn COOH_2.chk < RESP_2.in > RESP_2.out
得到的结果并不是加权的结果，而是和只用COOH_2.chk相同，观察输出文件RESP_2.out看到。

Atomic radii used:
Element:H    vdW radius (Angstrom): 1.200
Element:C    vdW radius (Angstrom): 1.500
Element:O    vdW radius (Angstrom): 1.400
Generating fitting points and calculate ESP for conformer 1
Running: /Volumes/Gaussian/g16/g16/formchk "COOH_2.chk" > /dev/null
Running: rm -f "COOH_2.fchk"
Number of MK fitting points used:    4621
Running: /Volumes/Gaussian/g16/g16/formchk "COOH_2.chk" > /dev/null
Running: /Volumes/Gaussian/g16/g16/cubegen 4 potential=SCF "COOH_2.fchk" ESPresult.cub -5 h < cubegenpt.txt > nouseout
Running: rm -f "COOH_2.fchk"
Generating fitting points and calculate ESP for conformer 2
Running: /Volumes/Gaussian/g16/g16/formchk "COOH_2.chk" > /dev/null
Running: rm -f "COOH_2.fchk"
Number of MK fitting points used:    4621
Running: /Volumes/Gaussian/g16/g16/formchk "COOH_2.chk" > /dev/null
Running: /Volumes/Gaussian/g16/g16/cubegen 4 potential=SCF "COOH_2.fchk" ESPresult.cub -5 h < cubegenpt.txt > nouseout
Running: rm -f "COOH_2.fchk"
Reloading the file when Multiwfn boots up...
Running: /Volumes/Gaussian/g16/g16/formchk "COOH_2.chk" > /dev/null
Running: rm -f "COOH_2.fchk"

**** Stage 1: RESP fitting under weak hyperbolic penalty
No charge constraint and equivalence constraint are imposed in this stage
Convergence criterion: 0.00000100
Hyperbolic restraint strength (a): 0.000500 Tightness (b): 0.100000
Iter: 1 Maximum charge variation: 0.4944458622
Iter: 2 Maximum charge variation: 0.0024299035
Iter: 3 Maximum charge variation: 0.0000029249
Iter: 4 Maximum charge variation: 0.0000000035
Successfully converged!

**** Stage 2: RESP fitting under strong hyperbolic penalty
Stage 2 of standard RESP fitting is skipped since no atom needs to be fitted

Center    Charge
   1(C ) 0.420070
   2(O ) -0.496189
   3(H ) 0.110723
   4(O ) -0.441652
   5(H ) 0.407048
Sum of charges: 0.000000
Conformer: 1 RMSE: 0.003191 RRMSE: 0.098380
Conformer: 2 RMSE: 0.003191 RRMSE: 0.098380
Weighted RMSE: 0.003191 Weighted RRMSE 0.098380

所有的文件可见附件
(, 下载次数 Times of downloads: 4)
用来生产针对构象一的RESP
Multiwfn COOH_1.chk < RESP1.in > RESP1.out
用来生产针对构象二的RESP
Multiwfn COOH_2.chk < RESP.in > RESP2.out
用来生产针对加权构象的RESP
Multiwfn COOH_2.chk < RESP_2.in > RESP_2.out
得到RESP_2.out的结果和RESP2.out相同。

作者
Author: sobereva 时间: 2019-11-12 02:11
用fchk结合官网上的Multiwfn最新版本再试，如果还不行，把fchk文件上传

作者
Author: wuzhiyi 时间: 2019-11-12 05:06

sobereva 发表于 2019-11-12 02:11
用fchk结合官网上的Multiwfn最新版本再试，如果还不行，把fchk文件上传

谢谢sob老师，用最新版的multiwfn发现问题仍然存在。
但在将文件转换为fch之后作为输入文件，问题解决。

作者
Author: sobereva 时间: 2019-11-12 11:02
看来是bug，每次调用formchk转换的都是COOH_2.chk
今天我会修正这个问题，在官网上更新

作者
Author: sobereva 时间: 2019-11-12 16:20
我已更新了官网上的Multiwfn 3.7(dev)修正了这个问题

作者
Author: wuzhiyi 时间: 2019-11-12 22:40

sobereva 发表于 2019-11-12 16:20
我已更新了官网上的Multiwfn 3.7(dev)修正了这个问题

谢谢sob老师

欢迎光临计算化学公社 (http://bbs.keinsci.com/)