« Back
Generating Unbiased Structural Alternatives for Automated Structure Verification
Poster Title: Generating Unbiased Structural Alternatives for Automated Structure Verification
Submitted on 02 Oct 2017
Author(s): Sergey Golotvin, Rostislav Pol, Mikhail Elyashberg, Dimitris Argyropoulos and Karim Kassam
Affiliations: Advanced Chemistry Development, Inc., 8 King Street East, Suite 107, Toronto, Ontario, Canada
This poster was presented at SMASH 2017 - Small Molecule NMR Conference
Poster Views: 825
View poster »

Poster Information
Abstract: Automated structure verification (ASV) using NMR data is gaining acceptance as a routine application for qualitative evaluation of large compound libraries produced by synthetic chemistry. A proposed structure is confirmed if it fits a number of conditions from 1D 1H NMR data [1] or from a combination of 1D-1H and HSQC data.

Although it may be easy to eliminate proposed structures that do not meet filtering requirements, a more difficult task is to guarantee the structure, even when all filtering requirement are met – as there is always a possibility that an isomeric structure fits the same NMR data better than the proposed structure. One way to decrease the possibility that false structures will pass, is to use as many NMR experiments as possible for structure validation (1D-13C, HMBC, etc.). Another method [2] is to bring in and simultaneously verify several structures isomeric to the proposed. This concurrent structure verification enables structural differences to be highlighted through gradual tightening of NMR prediction tolerances until only a single structure remains.

The spectral similarities of isomeric structures to the one proposed can mean that interpretation may be biased by the chemist’s expectations, and therefore result in false positives.

Here we present a method for generating alternative ‘unbiased’ structures for ASV, based on the structure generator [3] used in a CASE system. Using this approach we are able to generate all isomers that fit a particular NMR dataset without using any prior available knowledge. Practical aspects of this ‘unbiased’ verification are discussed and several examples are shown.

Summary: A new method, Unbiased Verification (UBV), to generate and explore all possible structural alternatives compatible with NMR datasets is presented. The method can be applied to any proposed structure that passes standard NMR verification and can potentially be falsely considered as positive. References: Golotvin, Sergey S., Vodopianov, Eugene, Pol, Rostislav, Lefebvre, Brent A., Williams, Antony J., Rutkowske, Randy D., Spitzer, Timothy D., Mag. Res. Chem. 45(10), 803-813, 2007.

Golotvin, Sergey S., Pol, Rostislav, Sasaki, Ryan R., Nikitina, Asya, Keyes, Philip, Mag. Res. Chem. 50(6), 429-435, 2012.

Elyashberg, Mikhail E, Williams, Antony J., Blinov, Kiril.A., “Contemporary computer-assisted approaches to molecular structure elucidation”, RSC, Cambridge, 2012
Report abuse »
Ask the author a question about this poster.
Ask a Question »

Creative Commons

Related Posters

Artificial Intelligence: An Accomplishment of Modern Health Care System
Ananya Choudhury

World on Fire
Emilio Sempris

Combining Imaging and OMICs: MALDI guided SpatialOMx probes proteomic mass profiles of breast cancer tumor subpopulations
Janina Oetjen 1 , Romano Hebeler 1 , Frédéric Dewez 2 , Corinna Henkel 1 , Benjamin Balluff 2 , Ron Heeren 2

Machine Learning Approach to Geometry Prediction in Cold Spray Additive Manufacturing
Daiki Ikeuchi

Evaluation of Laser Cleaning of Parchment, Wool and Featherwith High Performance Liquid Chromatography and AttenuatedTotal Reflection -Fourier Transform Infrared Spectroscopy
Stamatis C. Boyatzis, Eleni Ioakimoglou, Eleni Tziamourani, Efrosini Karantoni Ekaterini Malea, Stavroula Rapti, Paraskevi Pouli, Athanassia Papanikolaou, Kristalia Melessanaki, Georgios Panagiaris