« Meet Quindolinocryptotackiene | Main | Fringe Benefits and Knowledge Management »

October 25, 2007


Feed You can follow this conversation by subscribing to the comment feed for this post.

Philip Keyes

I found the question of accuracy of chemical shifts relative to automated collection of data quite relevant with respect to our own automated verification project. In fact, one of the early confounding problems was a simple induced error of about 0.12 ppm in proton NMR that was the result of low level impurities in medchem spectra that were presumed to be at 0 ppm by the default action of the SREF command on a Bruker spectrometer. These were possibly silica grease related etc. This occasional induced error would affect the dark regioning of the compounds residual lock solvent and also shift the actual peaks of the spectrum enough to siginificantly reduce the automated scoring. So not only does reported chemical shift errors create prediction verification problems, so do the automated referencing induced errors on the experimental spectra that are to be compared to the predictions. In order to reduce this induced shift error, we have opted to remove the SREF command from our proc_1d AU programs and instead rely purely on the lock solvent for referencing and of course appropriate adjustment and configuration of our edlock table.

Ryan Sasaki

Thanks for sharing your thoughts Phil, as well as the tip

Ryan Sasaki

For those interested, my colleague Kirill has pointed me to an article written by Michael Grzonka and Antony Davies entitled,

"Emperical Investigation on the Reproducibility of 13C NMR Shift Values"

It's an interesting read where the authors investigate how the polarity of solvents, referencing of samples, as well as faulty spectrometer and software configurations effect the reproducibility of experimental 13C NMR spectra.

Here's a link to the abstract:


Philip Keyes

The ability to handle solvent specific NMR prediction is a key element of realistic scoring. Currently, automation server version 10.05+, with specific and selective DB training, can handle only training DB selective solvent specific training (with scripting)... however, it certainly seems, that at some point in the future, universal solvent selective training will be a reality. Its been discussed for quite a few versions now but it seems close to coming to fruition. This should improve the quality of the verification scoring, as well as the the utility of predictions. Down the road , the next most important thing would be taking into account concentration and pH, but I think that would be quite a way off. It is still incumbent on the instrument managers to properly set up their own instrumentation to properly reference compounds. Two choices exist. Either mandatory reference compound (TMS) in lock solvents which can be bought that way ... or lock referencing on deuterium. For the convenience of our own chemists workflow, we choose the later since it aids in their own recovery of the sample... if they need to.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been saved. Comments are moderated and will not appear until approved by the author. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Comments are moderated, and will not appear until the author has approved them.

Your Information

(Name is required. Email address will not be displayed with the comment.)