Multiwfn forum

limmi

Member

Registered: 2023-11-23

Posts: 2

Missing reference parameter using HOMA

Dear Multiwfn community,

I am currently using Harmonic oscillator measure of aromaticity (HOMA) on a set of molecules and encountered "Missing reference parameter" errors, for some of the molecules. I suspect the problem is the utilization of ghost atoms. Here is one example output after starting the HOMA calculation:
"
Current reference bond length (Angstrom) and sigma parameters:
B -C : 1.4235 104.5070
B -N : 1.4020 72.0300
C -C : 1.3880 257.7000
C -N : 1.3340 93.5200
C -O : 1.2650 157.3800
C -P : 1.6980 118.9100
C -S : 1.6770 94.0900
N -N : 1.3090 130.3300
N -O : 1.2480 57.2100

Input indices of the atoms involved in the ring, e.g. 1,5,6,7,8,12
(Input q can return)
2, 1, 6, 7, 4, 3

Atom pair Contribution Bond length(Angstrom)
2(C ) -- 1(O ): -0.182123 1.348327
Error: Missing reference parameter for O -O
"

Best regards,

limmi

Last edited by limmi (2023-11-23 13:56:20)

sobereva

Tian Lu (Multiwfn developer)

From: Beijing

Registered: 2017-09-11

Posts: 1,768

Website

Re: Missing reference parameter using HOMA

Dear limmi,

Please upload your input file or send your input file to my E-mail, I can hardly answer this question without the input file for Multiwfn.

Best,

Tian

sobereva

Tian Lu (Multiwfn developer)

From: Beijing

Registered: 2017-09-11

Posts: 1,768

Website

Re: Missing reference parameter using HOMA

I have checked your .fch file, the atoms in the ring (according to bonding relationship) is 2,3,4,5,8,7, therefore you should input 2,3,4,5,8,7 in this interface, and result can be normally obtained.

Atom pair Contribution Bond length(Angstrom) 2(C ) -- 3(N ): -0.022401 1.371910 3(N ) -- 4(C ): -0.064812 1.398484 4(C ) -- 5(N ): -0.033910 1.380643 5(N ) -- 8(C ): -0.109995 1.418006 8(C ) -- 7(C ): -0.170641 1.451032 7(C ) -- 2(C ): -0.037877 1.358303 HOMA value is 0.560364

limmi

Member

Registered: 2023-11-23

Posts: 2

Re: Missing reference parameter using HOMA

Thanks a lot for your answer!

I am a bit surprised, because the atomindices I used [2, 1, 6, 7, 4, 3] are correct according to rdkit for this molecule OC(NC(N1)=O)=CC1=O

However using the atom indices from the fchk [2,3,4,5,8,7] I am getting different atoms in rdkit:

Edit: I think i found the problem: Rdkit's numbering starts from 0, while the numbering for the .fchk files starts from 1.

Best regards,

limmi

Last edited by limmi (2023-11-24 14:06:56)