A Statistical Approach To Identifying the Batch of Origin of Mixed-Meat Products Using DNA Profiles

摘要

Comparison of DNA samples at different points of a supply chain offers a powerful means of verifying tracing systems for primal cuts of meat. However, this approach is problematic for products made from ground (or mixed) meat because such products are typically made from an unknown (and random) number of unidentified animals. We present a statistical method that uses DNA profiles to verify or refute the contention that a particular mixed-meat product came from a particular manufacturing batch. This method involves randomly isolating a number of individual DNA samples (comprising an unknown number of individual genotypes) from the end product and comparing them with a set of DNA samples (also comprising an unknown number of individuals) that had been collected randomly before preparation of a manufacturing batch. Confidence levels are given for refuting spurious claims, and the development of optimum sampling strategies is discussed. The results are discussed in relation to batch verification of mixed-meat products in the food industry, with an emphasis on traceability issues.