Identification of essentially derived varieties obtained from biparental crosses of homozygous lines. III. AFLP data from maize inbreds and comparison with SSR data

Genetic distance (GD) estimates based on molecular markers are a preferred approach to estimate genetic conformity between putative essentially derived varieties (EDVs) and their initial varieties (IVs). The objectives of our study were to (1) estimate the variation in the marker-estimated parental contribution (p) to the genome of the progeny, (2) investigate the power of AFLP-based GD estimates for discriminating between homozygous lines with F-2, BC1, and BC2 relationships, and (3) compare AFLP data with SSR data from a companion study, as well as with theoretical and simulated results. In total, 104 European maize inbred lines comprising 66 triplets were analyzed with 20 AFLP primer combinations and 100 SSR primer pairs. Each triplet consisted of one F-2-, or BC1-derived progeny line and both parental lines. For lines with both F-2 and BC1 relationship, mean estimates of p based on AFLP (p(F2) = 0.46; p(BC1) = 0.70) or SSR data (p(F2) = 0.45; p(BC1) = 0.67) were lower than the respective expected values of 0.50 and 0.75. Considerable differences were detected between AFLP- and SSR-based mean GD estimates for unrelated inbred lines due to the lower degree of polymorphism of AFLPs compared with SSRs. With each marker system, the GD between progeny lines and parents was little affected by the variation in GD between the parents, particularly for lines with BC1 relationship. Substantial differences in Type I (alpha) and Type II (beta) errors were detected between flint and dent germplasm pools with different marker systems and when fixed EDV thresholds were considered. To counterbalance advantages and disadvantages of AFLP and SSRs, a complementary application of the two marker systems is suggested for discrimination between EDVs and independently derived varieties (IDVs).