QSARs - PowerPoint PPT Presentation

1 / 30
About This Presentation
Title:

QSARs

Description:

'Correlation between Hydrophobicity of Short-Chain Aliphatic Alcohols and Their ... I and Class II Compounds By Hydrophobicity and Hydrogen Bonding Descriptors. ... – PowerPoint PPT presentation

Number of Views:292
Avg rating:3.0/5.0
Slides: 31
Provided by: stude603
Category:

less

Transcript and Presenter's Notes

Title: QSARs


1
QSARs
  • Quantitative Structure-Activity Relationships and
    their Applications

Robin Hughes
2
QSARs
  • What Are They?
  • Why Do We Care?
  • How Are They Used?
  • Limitations
  • Conclusions / Whats Next?

3
What Are They?
  • Mathematical Models
  • Chemical Structure ?Biological Activity
  • Hydrophobic
  • Electronic
  • Steric

4
Why Do We Care?
  • Based on Parameters
  • Time
  • Cost
  • Predictive

5
How Are They Used?
  • Recent (Past 10yrs)
  • Pharmaceutical, Agricultural
  • Establishing Priorities
  • Estimation
  • Risk Assessment
  • International Decision-Making

6
Studies
  • Correlation between Hydrophobicity of
    Short-Chain Aliphatic Alcohols and Their Ability
    to Alter Plasma Membrane Integrity McKarns et
    al. (1996)
  • Classifying Class I and Class II Compounds By
    Hydrophobicity and Hydrogen Bonding Descriptors
    S. Ren (2002)
  • Structure- and Property- Activity Relationship
    Models of Microbial Toxicity of Organic Chemicals
    to Activated Sludge N. Nirmalakhandan,
    E.Egemen, C. Treviso, and S. Xu (1997)

7
Correlation between Hydrophobicity of Short-Chain
Aliphatic Alcohols and Their Ability to Alter
Plasma Membrane Integrity McKarns et al. (1996)
  • Lactate dehydrogenase (LDH)
  • Log P ( log Kow)

8
Study Design
  • 2 Models
  • LDH50
  • EC50
  • Almost Identical

9
(No Transcript)
10
(No Transcript)
11
(No Transcript)
12
Findings
  • LDH50 and EC50 are positively correlated with
    Hydrophobicity
  • Models are good predictors
  • Predict effects of untested chemicals

13
Classifying Class I and Class II Compounds By
Hydrophobicity and Hydrogen Bonding
Descriptors S. Ren (2002)
  • Class I Nonpolar Narcotic
  • Baseline Toxicity
  • Class II Polar Narcotic
  • Above Baseline Toxicity

14
Study Design
  • Nonlinear model
  • Discriminant Analysis
  • Descriptors Used
  • Log Kow
  • ELUMO, EHOMO
  • Q, Q-

15
Findings
  • Applied to 190 compounds
  • 8 Misclassifications
  • All 5 variables significant

16
Structure- and Property- Activity Relationship
Models of Microbial Toxicity of Organic Chemicals
to Activated Sludge
N. Nirmalakhandan, E.Egemen, C. Treviso, and S.
Xu (1997)
  • Concern
  • Industrialization ? SOCs in wastewater
  • Microorganisms used in treatment
  • What are threshold levels?

17
Study Design
  • 4 Models
  • IC50
  • Experimental Data 16 chemicals

18
QSAR / QPAR
  • Structure/Property
  • Authors define analyses dealing with
    octanol-water partition coefficients and aqueous
    solubility as QPARs
  • QSARs defined using solvatochromatic parameters
    (Linear Solvation Energy Relationships - LSER)
    and molecular connectivity indices (MCI)

19
Models
QSAR Models
  • LSER Model
  • VI, Instrinsic molar volume
  • p, Polarity/Polarizability
  • am, Hydrogen bond donor acidity
  • ßm, Basicity

20
Models
QSAR Models
  • MCI Model
  • Structural and Atomic Information
  • Express physical, chemical, and biological
    properties

21
Models
QPAR Models
  • Octanol-Water Partition Model
  • log P
  • Used calculated values for consistency

22
Models
QPAR Models
  • Aqueous Solubility Model
  • log S correlated with other indicators
  • Hypothesis toxicity may be directly correlated
    with S
  • Used experimental values

23
Findings
  • LSER Model
  • High statistical validity, but not the best
    toxicity predictor
  • Log S
  • No statistically significant correlation found
  • Both models above acceptable FE for
  • 50 of tested chemicals

24
Findings
  • log P and MCI Models
  • reliable and convenient
  • Parameters easily available
  • Further testing of utility suggested

25
Findings
  • Predicted vs Experimental
  • LC50
  • All 4 models

26
Findings
  • Factor of error
  • FEpredictive/measured
  • If FElt1 1/FE was used

27
Limitations
  • Activity-Specific
  • Experimental Error
  • Too Many Parameters
  • Validation
  • Projecting Beyond Sample Space

28
Conclusions
  • Similar Parameters
  • Still Secondary
  • Low-Cost (both time and )
  • Mechanism of Action
  • Still in development

29
Whats Next?
  • Databases
  • Predictive Capabilities
  • Prevention
  • Remediation?

30
Works Cited
  • Cronin, M. et al. (2003). Use of Quantitative
    Structure-Activity Relationships in International
    Decision-Making Frameworks to Predict Health
    Effects of Chemical Substances. In Journal of the
    National Institute of Environmental Health
    Sciences. (www.ehponline.org).
  • McKarns, S., et al. (1997). Correlation between
    Hydrophobicity of Short-Chain Aliphatic Alcohols
    and Their Ability to Alter Plasma Membrane
    Integrity. In Fundamental and Applied Toxicology,
    36 62-70.
  • Nirmalakhandan, N.,.Egemen, E, Treviso, C., and
    Xu, S. (1998). Structure- and Property- Activity
    Relationship Models of Microbial Toxicity of
    Organic Chemicals to Activated Sludge. IN
    Ecotoxicology and Enironmental Safety,
    39112-119.
  • Rand, G.M., Ed. Fundamentals of Aquatic
    Toxicology, 2nd Ed. Taylor and Francis.
    Philadelphia, PA, 1995.
  • Ren, S. (2002). Classifying Class I and Class II
    Compounds By Hydrophobicity and Hydrogen Bonding
    Descriptors . In Environmental Toxicology, 17
    415-423.

31
QSAR and Modeling Society
  • http//www.ndsu.nodak.edu/qsar_soc/aboutsoc/applic
    at.pdf
Write a Comment
User Comments (0)
About PowerShow.com