|
Activity Number:
|
223
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 10, 2004 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #301907 |
|
Title:
|
Acceptance Sampling Methods Extensions to Crop Seed Purity Testing for Transgenic Traits
|
|
Author(s):
|
Kirk M. Remund*+ and Jean-Louis Laffont and Deanne L. Wright
|
|
Companies:
|
Monsanto and Pioneer Génétique and Pioneer Hi-Bred International Inc.
|
|
Address:
|
800 North Lindbergh Blvd., St. Louis, MO, 63167,
|
|
Keywords:
|
acceptance sampling ; GMO testing ; assay errors ; DNA ; PCR
|
|
Abstract:
|
Acceptance sampling methods have been used for decades to reject or accept product lots before release to the consumer. Many industries have even abandoned the use of acceptance sampling methods because end-product testing is no longer needed. However, in crop seed testing for transgenic trait purity or impurity, these methods have proven very useful in recent years. Purity tests include testing for protein expression via ELISA methods and DNA sequences via polymerase chain reaction methods (PCR). This paper presents some useful extensions to the standard acceptance sampling methods based on the binomial distribution for seed purity testing. These extensions include modifications to allow purity characteristics to be measured on pools of many seeds rather than individual seeds. An approach to adjust acceptance sampling plans using qualitative methods by the assay false positive and false negative rates is presented. For quantitative PCR assays, standard acceptance sampling methods extensions are made to allow method and subsampling variability estimates from a variance components model to be coupled with binomial sampling variability in the operating characteristic curve.
|
