Posts Tagged neural net NMR
During my career I have had the opportunity to work with some of the best scientists. During my >10 year tenure at ACD/Labs I directly managed the NMR product line for a number of years and was involved with the development of both the NMR prediction and Computer Assisted Structure Elucidation software. Over the past couple of years since leaving ACD/Labs I have continued to co-author articles with two of my ex-colleagues, and still friends, Kirill Blinov and Mikhail Elyashberg. Three of these articles were released in one work. If you are interested in NMR prediction and CASE systems you might find these articles interesting.
The application of empirical methods of 13C NMR chemical shift prediction as a filter for determining possible relative stereochemistry
Mikhail E. Elyashberg, Kirill A. Blinov, Antony J. Williams
Published Online: Feb 10 2009 8:59AM
The reliable determination of stereocenters contained within chemical structures usually requires utilization of NMR data, chemical derivatization, molecular modeling, quantum-mechanical calculations and, if available, X-ray analysis. In this article we show that the number of stereoisomers which need to be thoroughly verified can be significantly reduced using NMR chemical shift calculations for the full stereoisomer set of possibilities using a fragmental approach based on HOSE codes. The applicability of this suggested method is illustrated using a series of complex chemical structures.
A systematic approach for the generation and verification of structural hypotheses
Mikhail Elyashberg, Kirill Blinov, Antony Williams
Published Online: Feb 5 2009 4:33AM
In this article we show that the most rational manner by which to create structural hypotheses is by the application of an expert system capable of deducing all potential structures consistent with the experimental spectral data. Empirical or quantum-mechanical (QM) NMR prediction methods are compared. It is shown that when an expert system is used the best structure(s) can be distinguished using either incremental or neural net (NN)-based NMR prediction algorithms.
Development of a fast and accurate method of 13C NMR chemical shift prediction
Available online 11 February 2009
Kirill A. Blinov, Yegor D. Smurnyy, Tatiana S. Churanova, Mikhail E. Elyashberg, Antony J. Williams
In this article we describe a fast and accurate method of 13C NMR chemical shift prediction. The high speed of chemical shift calculation described is achieved using a simple structure description scheme based on individual atoms rather than functional groups. The systematic choice of an appropriate encoding scheme and the usage of partial least squares regression on a large training set has resulted in a robust and fast algorithm. The approach provides accuracy comparable with other well known approaches but demonstrates accelerated calculation speeds of up to a thousand times faster.