Burkholderia pseudomallei: Dose Response Models

From QMRAwiki
Revision as of 20:20, 19 May 2012 by Sidgunawardena (talk | contribs) (Summary Data)
Jump to: navigation, search

Burkholderia mallei and pseudomallei

Author: Mark H. Weir
If you want to download this chapter in pdf format, please click here


Overview: Burkholderia and meliodosis

Melioidosis, which typically infects horses, mules and donkeys but can also present a potentially life threatening disease in humans as well, is caused by the bacterium Bukholderia pseudomallei (B. pseudimallei). The bacterium is present in numerous tropical regions such as Central and South America and Southeast Asia. Burkholderia mallei (B. mallei) is a close relative of B. pseudomallei both of which have been categorized as a B level bioterror agent by the CDC.



Summary Data

Brett, P.J. and Woods, D.E.(2000) Pathogenesis of and Immunity to Melioidosis Acta Tropica 74: 201-210

Miller, W.R., Pannel, I., Cravitz, I., Tanner, W.A., Rosebury, T. (1947) Studies on Certain Biological Characteristics of Melleomyces mallei and Malleomyces pseudomallei II. Virulence and Infectivity for Animals Journal of Bacteriology 55: 127-135


Experiment serial number Reference Host type Agent strain Route # of doses Dose units Response Best fit model Optimized parameter(s) LD50/ID50
18* [1] mouse KHW intranasal 5 CFU exponential k = 1E-04 6.92E+03
17 [1] mouse KHW intranasal 5 CFU exponential k = 1.04E-02 6.63E+01
21 [2] guinea pig W294 intraperitoneal 6 CFU death beta-Poisson α = 2.67E-01 , N50 = 2.55E+02 2.55E+02
*This model is preferred in most circumstances. However, consider all available models to decide which one is most appropriate for your analysis.

* Recommended Model

Since intranasal exposure is closer to inhalation which is the likeliest exposure route for humans, especially compared to intraperitoneal, the data sets and resulting models from the Brett and Woods (2000) article is preferred. Also given the improved fit between the two data sets experiment 18 is chosen as the recommended model.

a:
Exponential and betapoisson model.jpg

Optimization Output for experiment 18

C57BL/6 Mice KHW Strain Data [1]
Dose Infected Non-infected Total
150 0 6 6
450 1 5 6
1350 1 5 6
4050 3 3 6
12200 3 3 6


Goodness of fit and model selection
Model Deviance Δ Degrees
of freedom
χ20.95,1
p-value
χ20.95,m-k
p-value
Exponential 3.36 2.17 4 3.84
0.141
9.49
0.499
Beta Poisson 1.19 3 7.81
0.755
Exponential is preferred to beta-Poisson; cannot reject good fit for exponential.


Optimized parameters for the exponential model, from 10000 bootstrap iterations
Parameter MLE estimate Percentiles
0.5% 2.5% 5% 95% 97.5% 99.5%
k 1E-04 2.87E-05 4.15E-05 4.99E-05 1.89E-04 2.13E-04 2.70E-04
ID50/LD50/ETC* 6.92E+03 2.57E+03 3.26E+03 3.68E+03 1.39E+04 1.67E+04 2.41E+04
*Not a parameter of the exponential model; however, it facilitates comparison with other models.


Parameter histogram for exponential model (uncertainty of the parameter)
Exponential model plot, with confidence bounds around optimized model


Optimization Output for experiment 17

BALB/c Mice KHW Strain Data [1]
Dose Infected Non-infected Total
5 0 6 6
15 0 6 6
45 3 1 4
135 4 2 6
405 6 0 6


Goodness of fit and model selection
Model Deviance Δ Degrees
of freedom
χ20.95,1
p-value
χ20.95,m-k
p-value
Exponential 5.25 -0.000362 4 3.84
1
9.49
0.263
Beta Poisson 5.25 3 7.81
0.154
Exponential is preferred to beta-Poisson; cannot reject good fit for exponential.


Optimized parameters for the exponential model, from 10000 bootstrap iterations
Parameter MLE estimate Percentiles
0.5% 2.5% 5% 95% 97.5% 99.5%
k 1.04E-02 4.94E-03 5.96E-03 6.60E-03 1.92E-02 2.45E-02 2.45E-02
ID50/LD50/ETC* 6.63E+01 2.82E+01 2.82E+01 3.61E+01 1.05E+02 1.16E+02 1.40E+02
*Not a parameter of the exponential model; however, it facilitates comparison with other models.


Parameter histogram for exponential model (uncertainty of the parameter)
Exponential model plot, with confidence bounds around optimized model


Optimization Output for experiment 21

Guinea Pigs W294 Strain Data [2]
Dose Dead Survived Total
44 1 4 5
440 3 2 5
4400 4 1 5
44000 5 0 5
440000 5 0 5
4.4E+06 4 1 5


Goodness of fit and model selection
Model Deviance Δ Degrees
of freedom
χ20.95,1
p-value
χ20.95,m-k
p-value
Exponential 70.3 66.1 5 3.84
4.44e-16
11.1
8.93e-14
Beta Poisson 4.14 4 9.49
0.387
Beta-Poisson fits better than exponential; cannot reject good fit for beta-Poisson.


Optimized parameters for the beta-Poisson model, from 10000 bootstrap iterations
Parameter MLE estimate Percentiles
0.5% 2.5% 5% 95% 97.5% 99.5%
α 2.67E-01 3.36E-02 8.80E-02 1.16E-01 1.05E+01 3.45E+02 3.55E+03
N50 2.55E+02 4.80E-07 2.49E+00 1.48E+01 1.35E+03 1.80E+03 3.22E+03


Parameter scatter plot for beta Poisson model ellipses signify the 0.9, 0.95 and 0.99 confidence of the parameters.
beta Poisson model plot, with confidence bounds around optimized model


References

  1. 1.0 1.1 1.2 1.3 Brett, P.J. and Woods, D.E.(2000) Pathogenesis of and Immunity to Melioidosis Acta Tropica 74: 201-210
  2. 2.0 2.1 Miller, W.R., Pannel, I., Cravitz, I., Tanner, W.A., Rosebury, T. (1947) Studies on Certain Biological Characteristics of Melleomyces mallei and Malleomyces pseudomallei II. Virulence and Infectivity for Animals Journal of Bacteriology 55: 127-135