Predicting Offspring Performance in Hop (Humulus Lupulus L.) Using AFLP Markers

Abstract

Genetic information for male and female hop accessions is limited, hampering parental selection to maximize offspring productivity. Our objective was to determine whether amplified fragment length polymorphism (AFLP)-derived genetic distance (GDmm) estimates and coancestry estimates from pedigrees (GSped) between parental pairs could be used to predict offspring performance. GDmm estimates among 16 cultivars and 31 male accessions were used to create breeding pairs that were either distantly related (HGD) or closely related (LGD). Families were grown under field conditions in a randomized complete block design with three replicates; yield and vigor were recorded in 2006 and 2007. The HGD families had nearly significantly higher yields (LS means = 1.56 kg per plant versus 1.21 kg per plant; P = 0.056) and significantly higher vigor (LS means = 7.29 [ordinal scale 1–10] versus 6.54; P = 0.0001) than LGD families. GDmm estimates were significantly correlated with yield (r = 0.2438; P = 0.023) and vigor (r = 0.2435; P = 0.024). GDmm estimates were also correlated with midparent heterosis (MPH) for vigor (r = 0.236; P = 0.03) but were not correlated with yield-based MPH or specific combining ability estimates based on yield or vigor. GSped estimates were negatively correlated with vigor (r = −0.312; P = 0.018) and vigor-based estimates of MPH (r = −0.291; P = 0.01). The information from this study suggests that molecular-based GDmm estimates and, in some cases, GSped estimates can predict superior offspring in a known set of hop accessions.