Elsevier

Neural Networks

Volume 6, Issue 7, 1993, Pages 1023-1032
Neural Networks

Neural networks for the peak-picking of nuclear magnetic resonance spectra

https://doi.org/10.1016/S0893-6080(09)80012-9Get rights and content

Abstract

Peak-picking is the lowest-level task of the interpretation of two-dimensional, and multidimensional Nuclear Magnetic Resonance (NMR) spectra in general, for protein structure determination. It consists of individuating peaks on two-dimensional frequency spectra, for further elaboration. The performances of several feedforward artificial neural networks trained with back propagation with temperature on the task of peak-picking are compared. The best one averages less than an approximate 5% error on well-defined spectral regions. The performances of the network are comparable with those of a human expert; the consequences of this fact on the possibility of improving further the performance of the network are discussed.

References (16)

There are more references available in the full text version of this article.

Cited by (30)

  • Recent Advances in Computational Methods for Nuclear Magnetic Resonance Data Processing

    2013, Genomics, Proteomics and Bioinformatics
    Citation Excerpt :

    Expected properties of peak shapes, such as the symmetry property, were used to identify peaks. Since then, a variety of computational methods have been utilized, including peak-property-based methods [11,12], machine learning methods [13–16], and spectra-decomposition-based methods [17–19]. Recently, image processing techniques have been applied to the peak picking problem and they have demonstrated promising performance [20,21].

  • Chapter 19 Neural networks for 2D NMR spectroscopy

    1996, Data Handling in Science and Technology
View all citing articles on Scopus
View full text