OpenTox2011_Talk-Rydberg

Document Sample
OpenTox2011_Talk-Rydberg Powered By Docstoc
					Prediction of Cytochrome P450
Mediated Metabolism




Patrik Rydberg
Ph.D., Associate Professor
Department of Medicinal Chemistry
University of Copenhagen
90% of Drugs
are metabolized by P450s
Creates toxicity
            Channel
Enzyme




Active
Site



         Heme
         Group
?
Binding   Reactivity   Entropy
               methoxy
Semi-empirical data is model
unreliable




 Aliphatic carbons     Aromatic carbons
200+ Transition States
   All major reaction types




     ΔE
40 kJ/mol
Maxbondsi = 5           Maxbondsall = 7



       Accessibility = 5 7 = 0.7
SMARTCyp


 Score = Energy – 8*Accessibility
               Isoform Unspecific Metabolism

                      SMARTCyp    CypScore
100

                                 94
80               85

60    67

40


20


 0
           1       1-2            1-3
                  Rank
Database Problems
    False negatives




Image from: M. Hennemann et al., ChemMedChem, 2009, 4, 657-669
 False negatives – N-dealkylation




Clomipramine                    Chlorpromazine
 False negatives – N-dealkylation




Clomipramine                    Chlorpromazine
SMARTCyp Accuracy for Nine Isoforms




      83                               81
 79
            74   74   73                     74    75
                           67
                                 58




1A2   2A6   2B6 2C19 2C8   2C9   2D6   2E1   3A4   Any
SMARTCyp Accuracy for Nine Isoforms



      83                               81
79
            74   74   73                     74    75
                           67
                                 58




1A2   2A6   2B6 2C19 2C8   2C9   2D6   2E1   3A4   Any
Enrichment




53               54                          53
      51    49        51                           49
                           45          45
                                 37




1A2   2A6   2B6 2C19 2C8   2C9   2D6   2E1   3A4   Any
www.farma.ku.dk/smartcyp/
 SMARTCyp is Flexible



                          Other
Web Server
                         Software




                          Other
Web Server    SMARTCyp
                         Software
SMARTCyp Extended with RS-Predictor

      86         86   84   84    84                86
82                                     81    82
            77




1A2   2A6   2B6 2C19 2C8   2C9   2D6   2E1   3A4   Any
The Next Step:
Sensible Atomic Descriptors

15 vs. 150 descriptors
100

 90

 80

 70

 60

 50                                                                 150
 40                                                                 15
 30

 20

 10

  0
      1A2   2A6   2B6   2C8   2C9   2C19   2D6   2E1   3A4 Merged
Metabolite prediction in
ToxTree 2.5

http://toxtree.sourceforge.net/
Usage and Availability
www.farma.ku.dk/smartcyp
ToxTree

MOE
Pipeline Pilot

Bioclipse
METEOR
What do we need?
What does the
future hold?
Better and more understandable models
More knowledge of P450 metabolism




                                        for references
                          google ”smartcyp”
Acknowledgements

University of Copenhagen, Denmark
Lars Olsen & David Gloriam

Rensselaer Polytechnic Institute, Troy, USA
Curt Breneman, Jed Zaretzki & Charles Bergeron

Funding
The Alfred Benzon Foundation
The Danish Medical Research Council
Lhasa Limited
The National Institutes of Health
The Office of Naval Research

				
DOCUMENT INFO
Categories:
Tags:
Stats:
views:0
posted:3/11/2013
language:English
pages:32