Try ACD/Labs software.

https://www.acdlabs.com/products/spectrus/workbooks/nmr/

Dear, Ahmad, You can predict the spectra by number of commerically available software such as Chemoffice and Mestrelab Mnova. Alternately you can use the free online link below

https://nmrshiftdb.nmr.uni-koeln.de/portal

http://www.cheminfo.org/Spectra/NMR/Predictions/1H_Prediction/index.html

Thank you very much fou your answers.

I’ll try it

Predicting the spectrum is not the same as interpreting the results (although it may help in some cases). The problem is that errors in prediction of chemical shifts or coupling constants may give rise to big differences in spectral appearance. ACD/Labs and Mestrelab have both tackled this in their own ways with varying degrees of success. As yet, there is no foolproof way of automatically interpreting NMR data today.

You don't say if you are confirming or elucidating a chemical structure. The former is normally referred to as "Automated Structure Verification" (ASV), the latter, "Computer Aided Structure Elucidation" (CASE).

Link to Mestrelab is here:

https://mestrelab.com/

Thank you Dr. Hellerton

Take the help of topspin or mnova

How

MNOVA is a good software.

Thank you very much 🙏🏻

What kind of spectrum do you have? Is it a pure compound ? Then you can make a prediction and compare both spectra (predicted by software vs. experimental). If you have the NMR spectrum of a mixture of compounds then you can say which functional groups are shown by which peaks and are thus present in the compounds. The 'untargeted mixture analysis' of ACD Labs sounds good.

Thank you Carolina Tabares-Mafla 👍

MestReNova is best software for NMR extraction. If you need any doubts about mnove fell free to ask. I have more experience on MNOVA

There is a free python program available on Github to help with AI structure elucidation. I have never used it myself, but I have heard good things about it. It may be a good alternative if you don't have a Mestrenova license. If you are going to test it and it works, please post your experience here.

https://github.com/KristapsE/DP4-AI

Luc Alders - the python program may be free but it has dependencies on substantial other software (e.g. Gaussian, MacroModel, RDkit, ++), so it isn't a case of "download the python program and go".

I haven't tried it but my reading of the paper indicates that it has been designed more to identify the most likely stereoisomer from a set of isomers. There is some nice work in the paper, much of it to do with extraction of features from the NMR spectrum. I know that Jonathan Goodman (one of the authors) has done a lot of good work in this area over the last few years.

John Hollerton - I was unaware of the further dependencies. I guess I will have to test it myself when I have some spare time. Spend time to save time, so to say.

The answer depends on many factors. I was working for >10 years on structure verification in Mnova; it works fine up to a certain number of atoms (like, say, 40 proton and as many carbons) and depends on such complicating factors as the presence of many labile protons, multiple diastereo pairs of protons, presence of rotamers and dynamic effects, impurities, S/N, etc. Also, you should add HSQC to the list of your spectra, otherwise the info content of of the spectra that you have may be simply insufficient.

But structure verification is not structure elucidation! The latter is much harder and, in my opinion, has always multiple solutions, at least with "real" spectra and with no a-priori knowledge. The best you can do today is some kind of computer-aided structure elucidation (CASE, the discipline started by Prof.Elyashberg).

There are various approaches and products, lots of people are working on it, but it so far always needs a spectroscopist to do most of the job. The computer can be a valid "partner" but it cannot know all the helpful contraints on what possibly might be present in the sample.

Thank you very much, Dr. Stan Sykora