Search PubMed⌕ Search
Home/Archive/PubMed

PubMed · 10340431

Estimating the linear analytic range.

H B Lipman·J R Astles
View original source ↗Export metadata ↓

Abstract

Although many methods have been proposed to verify assay linearity, or to test for nonlinearity, there are few, if any, statistically sound methods to establish an assay's linear range. In this article, we propose a simple and statistically sound method for initially establishing an assay's linear range as a paradigm for standardization of manufacturers' claims of assay linearity. Simulations verify that the method outperforms using (Pearson's) coefficient of determination (r2) to estimate the linear range, consistently yielding estimates for the linear range which are more accurate. In addition, the method permits the addition of tolerance limits when a certain amount of nonlinearity is acceptable clinically.

Explore related subjects

Linear ModelsReference StandardsJournal Article

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

H B Lipman, J R Astles. 1999. Estimating the linear analytic range.. https://doi.org/10.1016/s0009-8981(98)00219-8

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2025

A robust transfer learning approach for high-dimensional linear regression to support integration of multi-source gene expression data.

Transfer learning aims to integrate useful information from multi-source datasets to improve the learning performance of target data. This can be effectively applied in genomics when we learn the gene associations in a target tissue, and data from other tissues can be integrated. However, heavy-tail distribution and outliers are common in genomics data, which poses challenges to the effectiveness of current transfer learning approaches. In this paper, we study the transfer learning problem under high-dimensional linear models with t-distributed error (Trans-PtLR), which aims to improve the estimation and prediction of target data by borrowing information from useful source data and offering robustness to accommodate complex data with heavy tails and outliers. In the oracle case with known transferable source datasets, a transfer learning algorithm based on penalized maximum likelihood and expectation-maximization algorithm is established. To avoid including non-informative sources, we propose to select the transferable sources based on cross-validation. Extensive simulation experiments as well as an application demonstrate that Trans-PtLR demonstrates robustness and better performance of estimation and prediction when heavy-tail and outliers exist compared to transfer learning for linear regression model with normal error distribution. Data integration, Variable selection, T distribution, Expectation maximization algorithm, Genotype-Tissue Expression, Cross validation.

Linear Models↗
PubMed2001

Noncompartmentally-based pharmacokinetic modeling.

OBJECTIVE: To present an overview of noncompartmentally-based modeling which is a modeling that makes use of systems analysis, predominantly linear systems analysis (LSA). FINDINGS: Fundamental elements of LSA presented from a linear operational viewpoint have a sound foundation in molecular stochastic independence (MSI). Powerful LSA procedures based on MSI presented such as convolution, deconvolution and disposition decomposition analysis (DDA) enable PK predictions and evaluations of drug input and delivery using models with simple general structures and few verifiable assumptions. DDA nonparametrically differentiates the unit impulse response (UIR) into generalized elimination and distribution functions. DDA applied in a linear and nonlinear context is central to many LSA procedures such as analytically exact direct deconvolution, nonparametric evaluation of drug elimination and distribution, steady state predictions, evaluation of mean time parameters for drug delivery and disposition (mean residence time, mean transit time, mean arrival time), relative tissue affinity (residence time coefficients), nonparametric exact clearance correction of UIR, and time variant convolution and deconvolution. The general response mapping operation procedure of LSA presented provides a powerful rational alternative to problematic structured modeling of multivariate PK systems. CONCLUSION: The wide arsenal of underutilized LSA-based kinetic analysis tools provide a rational, powerful alternative to traditional kinetic modeling.

Linear Models↗
PubMed2001

Chain length dependence of lipid partitioning between the air/water interface and its subphase: thermodynamic and structural implications.

We have investigated phosphatidylcholines with the same two saturated hydrocarbon chains of 12, 10 and 8 C-atoms. Langmuir trough data could be evaluated towards even small lipid subphase desorption when applying a novel approach that had recently been developed in our laboratory. The C12 lipid turned out to form a nearly insoluble monolayer with slight desorption only beyond 15 mN/m for an area/volume ratio around 1 cm(-1). Above 22 mN/m micellation in the subphase apparently terminates further accumulation in the interface forcing additionally added lipid to enter the bulk volume. A comparatively substantial increase of solubilization was observed for the C10 monolayer. When turning to the C8 lipid partitioning proved to take place in nearly equal parts. In that case, strong multimeric aggregation is indicated to occur in both the interfacial and the bulk volume domains. All the results are quantitatively discussed in the light of basic thermodynamic and structural considerations.

Linear Models↗