Electrode potentials for bioreductive agents from neural networks

J. J. Wolfe, J. D. Wright, C. A. Reynolds, A. C.G. Saunders

Research output: Contribution to journalArticle

12 Citations (Scopus)

Abstract

The one-electron electrode potentials at pH 7 have been predicted to an average accuracy of about 70 mV for a number of nitrobenzenes, nitrofurans and nitroimidazoles using a neural network. The inputs were the heat of formation and the free energy of hydration of both the nitroarene and its radical anion. The heats of formation were calculated using semiempirical molecular orbital methods; the free energies of hydration were calculated using a modified Born equation with additional semiempirical terms. Since these inputs can be calculated quickly, the neural network promises to be very useful in the design of molecules such as bioreductive agents where the electrode potential is of crucial importance. The success of the neural network in this problem implies that the errors, primarily in the semiempirical heat of formation, are systematic, and offers the hope that these may be corrected in future generations of the semiempirical methods.

Original languageEnglish
Pages (from-to)85-102
Number of pages18
JournalAnti-Cancer Drug Design
Volume9
Issue number2
Publication statusPublished - 1 Jan 1994
Externally publishedYes

Keywords

  • Bioreductive agents
  • Electrode potentials
  • Neural networks
  • Nitrobenzenes
  • Nitrofurans
  • Nitroimidazoles
  • Nitrothiophenes
  • Semiempirical molecular orbital calculations

ASJC Scopus subject areas

  • Biochemistry
  • Oncology
  • Biochemistry, Genetics and Molecular Biology(all)
  • Pharmacology
  • Drug Discovery
  • Organic Chemistry

Fingerprint Dive into the research topics of 'Electrode potentials for bioreductive agents from neural networks'. Together they form a unique fingerprint.

  • Cite this

    Wolfe, J. J., Wright, J. D., Reynolds, C. A., & Saunders, A. C. G. (1994). Electrode potentials for bioreductive agents from neural networks. Anti-Cancer Drug Design, 9(2), 85-102.