Loading...

Principal component analysis-ranking as a variable selection method for the simulation of 13C nuclear magnetic resonance spectra of xanthones using artificial neural networks

Jalali Heravi, M ; Sharif University of Technology | 2007

304 Viewed
  1. Type of Document: Article
  2. DOI: 10.1002/qsar.200630111
  3. Publisher: 2007
  4. Abstract:
  5. A Quantitative Structure-Property Relationship (QSPR) relating atom-based calculated descriptors to 13C NMR chemical shifts was developed to accurately simulate 13C NMR spectra of polyhydroxy and methoxy substituted dibenzo pyrons (xanthones). A dataset consisting of 35 xanthones was employed for the present analysis. A set of 132 topological, geometrical, and electronic descriptors representing various structural characteristics was calculated for each of 497 unique carbon atoms in the dataset. Principal Component Analysis (PCA)-ranking and Artificial Neural Networks (ANNs) were used as descriptor selection and model building methods, respectively. Analyses of the results revealed a correlation coefficient and Root Mean Square Error (RMSE) of 0.998 and 1.42 ppm, respectively, for the prediction set. © 2007 Wiley-VCH Verlag GmbH & Co. KGaA
  6. Keywords:
  7. Carbon ; Xanthone derivative ; Artificial neural network ; Carbon nuclear magnetic resonance ; Principal component analysis ; Priority journal ; Proton nuclear magnetic resonance ; Quantitative structure activity relation
  8. Source: QSAR and Combinatorial Science ; Volume 26, Issue 6 , 2007 , Pages 764-772 ; 1611020X (ISSN)
  9. URL: https://onlinelibrary.wiley.com/doi/10.1002/qsar.200630111