ASSAYS FOR RECOMBINANT PROTEINS - A PROBLEM IN NONLINEAR CALIBRATION

Quantification of protein levels in biological matrices such as serum or plasma frequently relies on the techniques of immunoassay or bioass ay. The relevant statistical problem is that of non-linear calibration , where one estimates analyte concentration in an unknown sample from a calibration curve fit to known standard concentrations. This paper d iscusses a general framework for calibration inference, that of the no n-linear mixed effects model. Within this framework, we consider two i ssues in depth: accurate characterization of intra-assay variation, an d the use of empirical Bayes methods in calibration. We show that prop er characterization of intra-assay variability requires pooling of inf ormation across several assay runs. Simulation work indicates that use of empirical Bayes methods may afford considerable gain in efficiency ; one must weigh this gain against practical considerations in the imp lementation of Bayesian techniques. We illustrate the methods discusse d using a cell-based bioassay for the recombinant hormone relaxin.