Sanz Hector, Aponte John J, Harezlak Jaroslaw, Dong Yan, Ayestaran Aintzane, Nhabomba Augusto, Mpina Maxmillian, Maurin Obiang Régis, Díez-Padrisa Núria, Aguilar Ruth, Moncunill Gemma, Selidji Todagbe Agnandij, Daubenberger Claudia, Dobaño Carlota, Valim Clarissa
ISGlobal, Barcelona Ctr. Int. Health Res. (CRESIB), Hospital Clínic - Universitat de Barcelona, Barcelona, Spain.
Department of Biostatistics, Indiana University Fairbanks School of Public Health, Indianapolis, Indiana, United States of America.
PLoS One. 2017 Nov 14;12(11):e0187901. doi: 10.1371/journal.pone.0187901. eCollection 2017.
Multiplex bead-based immunoassays are used to measure concentrations of several analytes simultaneously. These assays include control standard curves (SC) to reduce between-plate variability and normalize quantitation of analytes of biological samples. Suboptimal calibration might result in large random error and decreased number of samples with analyte concentrations within the limits of quantification. Suboptimal calibration may be a consequence of poor fitness of the functions used for the SC, the treatment of the background noise and the method used to estimate the limits of quantification. Currently assessment of fitness of curves is largely dependent on operator and that may add additional error. Moreover, there is no software to automate data managing and quality control. In this article we present a R package, drLumi, with functions for managing data, calibrating assays and performing quality control. To optimize the assay the package implements: i) three dose-response functions, ii) four approaches for treating background noise and iii) three methods for estimating limits of quantifications. Other implemented functions are focused on the quality control of the fitted standard curve: detection of outliers, estimation of the confidence or prediction interval, and estimation of summary statistics. With demonstration purpose, we apply the software to 30 cytokines, chemokines and growth factors measured in a multiplex bead-based immunoassay in a study aiming to measure correlates of risk or protection from malaria of the RTS,S malaria vaccine nested in the Phase 3 randomized controlled trial of this vaccine.
基于微珠的多重免疫测定法用于同时测量几种分析物的浓度。这些测定包括对照标准曲线（SC），以减少板间变异性并使生物样品中分析物的定量标准化。校准不佳可能会导致较大的随机误差，并减少分析物浓度在定量限内的样品数量。校准不佳可能是由于用于SC的函数拟合不佳、背景噪声处理以及用于估计定量限的方法所致。目前，曲线拟合优度的评估在很大程度上依赖于操作人员，这可能会增加额外的误差。此外，没有软件来自动进行数据管理和质量控制。在本文中，我们展示了一个R包drLumi，它具有数据管理、测定校准和质量控制的功能。为了优化测定，该包实现了：i）三种剂量反应函数，ii）四种处理背景噪声的方法，以及iii）三种估计定量限的方法。其他实现的功能集中在拟合标准曲线的质量控制上：异常值检测、置信区间或预测区间估计以及汇总统计量估计。为了演示目的，我们将该软件应用于在一项研究中基于微珠的多重免疫测定中测量的30种细胞因子、趋化因子和生长因子，该研究旨在测量RTS,S疟疾疫苗疟疾风险或保护相关性，该疫苗嵌套在该疫苗的3期随机对照试验中。
