Q2. What have the authors stated for future works in "Bayesian analysis of serial dilution assays" ?
6. 3 Further Work An important direction of future research is to study which aspects of the curves vary between samples and which are stable, to allow the possibility of more accurate calibration. The model can potentially be improved in various ways, most notably by generalizing the function ( 1 ) of expected measurements. This represents a problem with both classical and Bayes estimates, and the authors suspect it is the reason why estimates from dilution assays are in practice much more variable than would be suggested by even the classical estimates in Figure 3.
Q3. What is the way to improve the accuracy of the assay?
By yielding more accurate estimates and quantifying inferential uncertainties (especially in the cases previously deemed outside detection limits), the Bayesian approach sets the stage for more systematic studies of model and design innovations, which the authors hope will lead to an even broader extension of the range of concentrations to which assays can be applied.
Q4. What are the weights for each unknown sample?
The weights depend on the unknown parameters θ, β, α, and so when the authors are fitting the model, the authors compute the set of weights for each of the unknown samples and normalize each set to sum to 1, for each posterior simulation draw.
Q5. What is the common design for estimating the concentrations of compounds in biological samples?
A common design for estimating the concentrations of compounds in biological samples is the serial dilution assay, in which measurements are taken at several different dilutions of a sample, giving several opportunities for an accurate measurement.
Q6. What is the advantage of the Bayesian approach?
The second advantage of the Bayesian approach is that it can incorporate several sources of variation without requiring point estimation or linearization, either of which can cause uncertainties to be underestimated in this nonlinear errors-in-variables model (Davidian and Giltinan, 1995; Dellaportas and Stephens, 1995).
Q7. What is the standard method used to estimate the concentrations of unknown samples?
the standards data are used to estimate the curve relating concentrations to measurements—typically assumed to be a four-parameter logistic function—using least squares.
Q8. What is the purpose of the hierarchical distribution for unknown concentrations?
When using the model to estimate unknown concentrations θ1, . . . , θJ , the authors fit a hierarchical model of the form,log θj ∼ N ( µθ, σ 2 θ ) , for j = 1, . . . , J.
Q9. What is the definition of a dilution assay?
Assays are performed using microtiter plates (for example, see Table 1) that contain two sorts of data: unknowns, which are the samples to be measured and their dilutions; and standards, which are dilutions of a known compound, used to calibrate the measurements.
Q10. What is the posterior median of the calibration curve?
The posterior median estimates of the parameters of the calibration curve are β̂1 = 14.8 (with a posterior 50% interval of [14.7, 15.0]), β̂2 = 94.3[89.8, 99.0], β̂3 = 0.048[0.044, 0.052], and β̂4 = 1.41[1.37, 1.46].
Q11. How do the authors use the weights to understand the information in existing data?
5.4 Using the Weights to Understand the Information in Existing DataFor any given unknown sample, the authors now have a weight for each measurement, and these can be normalized to sum to 1.
Q12. How can the authors estimate unknown concentrations using Bayesian methods?
The authors have also programmed the inference using the Metropolis algorithm (see, e.g., Gilks, Richardson, and Spiegelhalter, 1996) directly in R.Section 3.2 shows how Bayesian methods allow estimation of unknown concentrations with reasonable accuracy using much less data than needed by conventional methods, even with measurements that at first appear to be outside “detection limits.”
Q13. How many samples are available for the estimation of unknown concentrations?
In Section 4, the authors show that it is possible to obtain accurate estimates of concentrations varying by a factor of 46 or 47 (that is, ranging from sample 1 to sample 7 or 8 in Figure 5) and reasonable estimates over the entire range of 49, or 260,000.