|Van Vleck, Lloyd|
Submitted to: Animal Genetics
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/30/2001
Publication Date: 8/7/2001
Citation: Rocha, J.L., Pomp, D., Van Vleck, L.D., Nielsen, M.K. 2001. Predictors of marker-informativeness for an outbred F2 design. Animal Genetics. 32:365-370. Interpretive Summary: Genomic analyses to locate loci which affect quantitative traits (QTL) require information on marker alleles at other known chromosome locations. Some studies use markers from two parent lines and the F-2 progeny. Because parent lines are generally not homozygous at all loci in animals, not all marker alleles are informative (provide knowledge of QTL location). In this sstudy, two computer programs were developed to predict marker informativeness. One program assumed random mating (using marker frequencies) between the parent lines. The other program used the actual markers for individual parents. Actual markers of parents can be used to maximize marker information. In this study, marker information for location of QTL was improved from 15 to 20% by optimizing the matings based on parent markers for one chromosome with 28 markers.
Technical Abstract: Availability of dense marker maps leads to the need for criteria for marker-selection. Generalization of the Polymorphism Information Content (PIC) to represent marker-informativeness (MI) for a 3-generation F-2 design requires that two additional sources of non-informativeness be added to the PIC formula: the probability of matings between like-heterozygous F- -1 individuals, one of which is non-informative; and that of matings betwee like-heterozygous F-1 individuals, which are both fully informative but where line of origin of the same alleles is reciprocal. Two computer programs to predict MI were developed for an F-2 originating from two divergent selection lines of outbred mice (F-.2). A total of 403 markers had been genotyped for F-0 grandparents (n=31), and 14 markers had also been genotyped in the complete pedigree including 559 F-2. One program (RM) was based on assumptions of random-mating, while the other (PED) accounted for the pedigreed mating structure. For the 403 markers, the correlation between MI from RM and PED was .95, and the average deviation between the two predictions was .005 MI units (MI ranges from 0 to 1). Correlations between predicted and realized MI for 14 fully genotyped markers were .97 for PED and .94 for RM, while the corresponding averages of deviations between predicted and actual values were .01 and .04, respectively. Absolute deviations from realized MI never exceeded .09 and .16 for PED and RM, respectively. Simulated optimization of the mating system to maximize average MI of 28 markers on one chromosome led to improvements in the range of 15-20% average MI (.07-.09 MI units).