« Note from an NMRShiftDB Contributor | Main | N15 NMR Prediction Validation and Lighthouses »

June 27, 2007

Comments

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

Jeff Seymour

A few words to the community in order to clarify our definitions:

A "RECORD" within Modgraph's NMRPredict databases consists of the following information:

• Structure including stereochemical information
• Assigned shiftvalues
• Experimental conditions like solvent, temperature, techniques used for signal assignment
• Couplings, relaxation times, intensities, etc.
• Literature citation
• Remark field

The complete installation of all CNMR databases available within NMRPredict holds 345,308 records. A compound like 'benzene' has been measured under various conditions (e.g. Solvent: 'neat', CDCL3, DMSO-D6 ... to HF:TaF5) - this is counted as a 'record'. It is correct that these 56 benzenes are counted as 56 records, because every record contributes a new bit of information. For developing a solvent specific prediction scheme we believe it is absolutely necessary to have experimental data of IDENTICAL compounds, but run in different solvents. This very example of Benzene has been mentioned on our website and has been in our help files for many years - http://www.modgraph.co.uk/version3/product_nmr_help_solvents.htm

In order to satisfy Ryan's interest in the number of DIFFERENT STRUCTURES within our 345,308 records, here it is: 289,543 DIFFERENT STRUCTURES in 345,308 RECORDS

It is interesting to see that Ryan's blog from June 27th, 2007 comes up with a question which has already been answered by Wolfgang Robien on his web site on June 11th, 2007 - http://nmrpredict.orc.univie.ac.at/csearchlite/How_large_should_it_be.html
On this page the fundamental question about information gain related to the size of the knowledge base has been analyzed very carefully using exactly the C13NMR Predict databases with their 345,308 records.

Identical structure search:

When predicting the chemical shift values for a query structure, an identical structure search is started too. It is our INTENTION to display all stereoisomers of the query - we can easily change that on a customer's request. From our point of view it is valuable to have information about all
stereoisomers in one single table.

Quality of predictions:

Of course customers are really interested in how accurately a prediction program can predict THEIR molecules - not a collection of external data such as NMRShiftDB. We have always encouraged potential customers to try NMRPredict themselves and predict their own molecules and have made NMRPredict available for a free 45 day trial for many years. Go to the Mestrelab Research website - http://www.mestrec.com - and either download Mnova with the included NMRPredict Desktop plug-in or download the NMRPredict client and try your predictions on-line.

Ryan Sasaki

Jeff,

Thanks for clearing that information up. As I mentioned in my original posting, compiling multiple sources for one compound is extremely important. I just happen to believe that all of those sources should be compiled and counted as one record when talking about databases that serve as the backbone for NMR prediction. Structural diversity is the key.

I am glad you pointed out this information and provided a more relevant number to share with the public.

Dissertation Structure

Blogs are so informative where we get lots of information on any topic. Nice job keep it up!!

Verify your Comment

Previewing your Comment

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

Working...
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.

Working...

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.)