Multiwfn forum

saurav01

Member

Registered: 2021-06-29

Posts: 4

Multiwfn crashes while calculating RESP charge fitting

Hi,

I calculated pop=mk for a large enzyme cofactor in gaussian and then trying to do the RESP charge fitting. However as soon as I submit my gaussian log file the program crashes. Is there anyway to solve it.
please let me know if I need to attach my .fchk and .log files and how can I do that!
Thanks.

-Saurav

sobereva

Tian Lu (Multiwfn developer)

From: Beijing

Registered: 2017-09-11

Posts: 1,639

Website

Re: Multiwfn crashes while calculating RESP charge fitting

See http://sobereva.com/wfnbbs/viewtopic.php?id=521 on how to attach file.

If your file is larger than 2MB even after compressing, please send the files to my E-mail, I will check.

Also, please show me all commands you inputted in Multiwfn, so that I can reproduce your problem.

saurav01

Member

Registered: 2021-06-29

Posts: 4

Re: Multiwfn crashes while calculating RESP charge fitting

I mailed you the .fchk and .log files obtained from Gaussian 16 calculation. The steps I followed are outlined as below
>mcm_cofactor......fchk [Enter]
----files loaded successfully-----
----OPTIONS------------
>7 [Enter]
>18 [Enter]
>8 [Enter]
>1 [Enter]
------name of the file with geometry----------
>mcm_cofactor......log [Enter]
------CRASHHHHHHHHH------------

sobereva

Tian Lu (Multiwfn developer)

From: Beijing

Registered: 2017-09-11

Posts: 1,639

Website

Re: Multiwfn crashes while calculating RESP charge fitting

I have checked your files. As explicitly mentioned in Section 3.9.16.2 of Multiwfn manual, you must use IOp(6/33=2) keyword, otherwise ESP of fitting points will not be printed in Gaussian output file, and thus Multiwfn cannot load them.

I also strongly suggest removing all redundant keywords from your Gaussian input file. My recommended keywords: #P BP86/6-31g(d,p) pop=(MK,readradii) Guess=read IOp(6/33=2,6/42=6)

By the way, I suggest using TPSSh instead of the BP86 you originally used. According to my recent benchmark, ESP quality of TPSSh is evidently better than any pure functional including BP86. TPSSh is also very suitable for systems containing transition metals.

saurav01

Member

Registered: 2021-06-29

Posts: 4

Re: Multiwfn crashes while calculating RESP charge fitting

Hi,

I followed your recommendation and got RESP charges. Thanks a lot. However I had some minor questions for you
1) which functional is better for RESP charges if pure organic ligands (no transition metals) are involved - TPSSh or B3LYP?
2) How do you use MultiWfn to get RESP charges for chromophores as is found in Green Fluorescent Protein (GFP) ( I am sure you aware of the method of use of P2N file to restrict the charges on only chormophore-proper while at the same time capping the chromophore with two methyl groups whose charges are not calculated for example in RED Tools algorithm)??
Thanks.

sobereva

Tian Lu (Multiwfn developer)

From: Beijing

Registered: 2017-09-11

Posts: 1,639

Website

Re: Multiwfn crashes while calculating RESP charge fitting

Hi,

I followed your recommendation and got RESP charges. Thanks a lot. However I had some minor questions for you
1) which functional is better for RESP charges if pure organic ligands (no transition metals) are involved - TPSSh or B3LYP?
2) How do you use MultiWfn to get RESP charges for chromophores as is found in Green Fluorescent Protein (GFP) ( I am sure you aware of the method of use of P2N file to restrict the charges on only chormophore-proper while at the same time capping the chromophore with two methyl groups whose charges are not calculated for example in RED Tools algorithm)??
Thanks.

(1) Among all my tested functionals (except for double-hybrid functionals), wB97M-V performs best (you can use ORCA code to use this functional), the ESP quality of TPSSh is slightly better than B3LYP even for purely organic systems. However, B3LYP is also a satisfactory choice.

(2) I am not familiar with RED tools (specifically, I know there is such a program, but this program is really too complicated, this is one of reasons why I intended to develop RESP code in Multiwfn to fully replace it) and I've never heard of P2N file. In the RESP module of Multiwfn, you can easily set charge constraint condition so that the the sum of RESP charges of selected atoms exactly equals to a specific value. Please carefully check Section 4.7.7 of Multiwfn manual, there are relevant examples.