Journal of Computer-Aided Molecular Design

, Volume 8, Issue 6, pp 635–652

Compass: A shape-based machine learning tool for drug design

  • Ajay N. Jain
  • Thomas G. Dietterich
  • Richard H. Lathrop
  • David Chapman
  • Roger E. CritchlowJr.
  • Barr E. Bauer
  • Teresa A. Webster
  • Tomas Lozano-Perez
Research Papers

Summary

Building predictive models for iterative drug design in the absence of a known target protein structure is an important challenge. We present a novel technique, Compass, that removes a major obstacle to accurate prediction by automatically selecting conformations and alignments of molecules without the benefit of a characterized active site. The technique combines explicit representation of molecular shape with neural network learning methods to produce highly predictive models, even across chemically distinct classes of molecules. We apply the method to predicting human perception of musk odor and show how the resulting models can provide graphical guidance for chemical modifications.

Key words

Automated prediction QSAR Molecular shape Ligand binding Molecular recognition 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© ESCOM Science Publishers B.V 1994

Authors and Affiliations

  • Ajay N. Jain
    • 1
  • Thomas G. Dietterich
    • 1
    • 2
  • Richard H. Lathrop
    • 1
    • 3
  • David Chapman
    • 1
  • Roger E. CritchlowJr.
    • 1
  • Barr E. Bauer
    • 1
  • Teresa A. Webster
    • 1
    • 4
  • Tomas Lozano-Perez
    • 1
    • 3
  1. 1.Arris Pharmaceutical CorporationSouth San FranciscoU.S.A.
  2. 2.Computer Science DepartmentOregon State UniversityCorvallisU.S.A.
  3. 3.Artificial Intelligence LaboratoryMassachusetts Institute of TechnologyCambridgeU.S.A.
  4. 4.Computer Science DepartmentStanford UniversityStanfordU.S.A.

Personalised recommendations