A comparison of calibration methods for quantifictation of Basmati and non-Basmati rice using microsatellite analysis

Alison Colyer (1) , Roy Macarthur (2) , Joy Lloyd (2) , Hez Hird (2)

1 Statistics
2 Central Science Laboratory

The accurate quantitation of non-Basmati rice in Basmati rice is central to the successful prosecution of adulteration, where non-basmati rice has been substituted for Basmati rice. The current method and three alternatives of constructing calibration curves for the measurement of non-Basmati rice in Basmati rice using microsatellite analysis were investigated. The methods compared involved power regression, linear regression (with and without log10 transformation) and hyperbolic regression of the ratio of Basmati to non-Basmati peak areas. Assessments were made using the error uncertainty, standard error at the agreed limit of adulteration and 95% confidence intervals for five example data sets. The linear regression of the ratio of peak areas onto the ratio of content proportions was found to give the most precise calibration and thus enhanced quantification of the level of adulteration of Basmati rice with non-Basmati rice.

Mots clés: Life Sciences

Alison Colyer, Roy Macarthur, Joy Lloyd, Hez Hird. A comparison of calibration methods for quantifictation of Basmati and non-Basmati rice using microsatellite analysis. Food Additives and Contaminants, 2008, 25 (10), pp.1189-1194. ⟨10.1080/02652030802040141⟩. ⟨hal-00577373⟩