On small sample Bayesian inference and sequential design for quantal response curves

AbstractElectrochemical immunosensors enable rapid analyte quantification in small sample volumes, and have been demonstrated to provide high sensitivity and selectivity, simple miniaturization, and easy sensor production strategies. As a point-of-care (POC) format, user-friendliness is equally important and most often not combinable with high sensitivity. As such, we demonstrate here that a sequence of metal oxidation and reduction, followed by stripping via differential pulse voltammetry (DPV), provides lowest limits of detection within a 2-min automatic measurement. In exchanging gold nanoparticles (AuNPs), which dominate in the development of POC sensors, with silver nanoparticles (AgNPs), not only better sensitivity was obtained, but more importantly, the assay protocol could be simplified to match POC requirements. Specifically, we studied both nanoparticles as reporter labels in a sandwich immunoassay with the blood protein biomarker NT-proBNP. For both kinds of nanoparticles, the dose-response curves easily covered the ng∙mL−1 range. The mean standard deviation of all measurements of 17% (n ≥ 4) and a limit of detection of 26 ng∙mL−1 were achieved using AuNPs, but their detection requires addition of HCl, which is impossible in a POC format. In contrast, since AgNPs are electrochemically less stable, they enabled a simplified assay protocol and provided even lower LODs of 4.0 ng∙mL−1 in buffer and 4.7 ng∙mL−1 in human serum while maintaining the same or even better assay reliability, storage stability, and easy antibody immobilization protocols. Thus, in direct comparison, AgNPs clearly outperform AuNPs in desirable POC electrochemical assays and should gain much more attention in the future development of such biosensors.

Download Full-text

Bayesian Inference for Quantal Response Equilibrium in Normal-Form Games

SSRN Electronic Journal ◽

10.2139/ssrn.3748586 ◽

2020 ◽

Author(s):

James R. Bland

Keyword(s):

Bayesian Inference ◽

Normal Form ◽

Quantal Response ◽

Quantal Response Equilibrium ◽

Normal Form Games

Download Full-text

Bayesian jackknife empirical likelihood

Biometrika ◽

10.1093/biomet/asz031 ◽

2019 ◽

Vol 106 (4) ◽

pp. 981-988

Author(s):

Y Cheng ◽

Y Zhao

Keyword(s):

Bayesian Inference ◽

Empirical Likelihood ◽

Small Sample ◽

New Approach ◽

U Statistics ◽

Posterior Inference ◽

Jackknife Empirical Likelihood ◽

General Tool ◽

Control Study ◽

Better Than

Summary Empirical likelihood is a very powerful nonparametric tool that does not require any distributional assumptions. Lazar (2003) showed that in Bayesian inference, if one replaces the usual likelihood with the empirical likelihood, then posterior inference is still valid when the functional of interest is a smooth function of the posterior mean. However, it is not clear whether similar conclusions can be obtained for parameters defined in terms of $U$-statistics. We propose the so-called Bayesian jackknife empirical likelihood, which replaces the likelihood component with the jackknife empirical likelihood. We show, both theoretically and empirically, the validity of the proposed method as a general tool for Bayesian inference. Empirical analysis shows that the small-sample performance of the proposed method is better than its frequentist counterpart. Analysis of a case-control study for pancreatic cancer is used to illustrate the new approach.

Download Full-text