02267nas a2200313   4500000000100000008004100001260001300042653001300055653001100068653002000079653003800099653002300137653001100160653001200171653002500183653000900208653002000217653002400237100001400261700001400275700001200289700001300301700001100314245006300325300001200388490000700400520153200407022001401939       2006                            d  c2006 Dec10aBiometry10aFemale10aGenetic Linkage10aGenetic Predisposition to Disease10aGenomic Imprinting10aHumans10aleprosy10aLikelihood Functions10aMale10aModels, Genetic10aModels, Statistical1 aVincent Q1 aAlcaïs A1 aAlter A1 aSchurr E1 aAbel L00aQuantifying genomic imprinting in the presence of linkage.  a1071-800 v623 a<p>Genomic imprinting decreases the power of classical linkage analysis, in which paternal and maternal transmissions of marker alleles are equally weighted. Several methods have been proposed for taking genomic imprinting into account in the model-free linkage analysis of binary traits. However, none of these methods are suitable for the formal identification and quantification of genomic imprinting in the presence of linkage. In addition, the available methods are designed for use with pure sib-pairs, requiring artificial decomposition in cases of larger sibships, leading to a loss of power. We propose here the maximum likelihood binomial method adaptive for imprinting (MLB-I), which is a unified analytic framework giving rise to specific tests in sibships of any size for (i) linkage adaptive to imprinting, (ii) genomic imprinting in the presence of linkage, and (iii) partial versus complete genomic imprinting. In addition, we propose an original measure for quantifying genomic imprinting. We have derived and validated the distribution of the three tests under their respective null hypotheses for various genetic models, and have assessed the power of these tests in simulations. This method can readily be applied to genome-wide scanning, as illustrated here for leprosy sibships. Our approach provides a novel tool for dissecting genomic imprinting in model-free linkage analysis, and will be of considerable value for identifying and evaluating the contribution of imprinted genes to complex diseases.</p>  a0006-341X