Apomixis, or asexual seed development, if introduced into crop species, has the potential to greatly improve global food production. Towards this goal, this study focused on uncovering the genetic mechanisms that control the parthenogenesis step within apomixis whereby fertilisation is avoided. In the model apomict, Hieracium praealtum (Asteraceae), parthenogenesis is controlled by the LOSS OF PARTHENOGENESIS (LOP) locus. Previous research showed that in addition to genomic copies of candidate genes at LOP, the genome has at least three other copies referred to as alternative alleles. The main goal of this study was to investigate four candidate genes, Genes B, X, H and Y, at LOP by generating segregation data of the alternative alleles. BAC clones containing alternative allele sequences were identified and Roche 454 pyrosequenced. These sequences were used to design alternative allele specific primers for genotyping two Hieracium praealtum polyhaploid populations (~ 300 plants). Four major conclusions were drawn from this study. First, the alternative alleles were in fact acting like alleles to the LOP alleles of Genes B, X and Y. Second, allelic sequence divergence (ASD) of the LOP alleles of Genes B and X relative to the alternative alleles, indicated a recent and separate evolutionary history. Third and, unexpectedly, recombination was detected at the LOP locus, in contrast to other apomixis loci reported in the literature. Furthermore, Gene B was found to be very closely associated with parthenogenesis in the polyhaploid population indicating that it may be essential to parthenogenesis and therefore requires further investigation. On the other hand, the absence of Genes X, Y and H, due to recombination, had no impact upon parthenogenesis. Fourth, the sequence data suggested that the LOP and alternative alleles originated from a shared common allele ancestor. It is hoped that these findings have made a significant contribution towards the future goal of introducing apomixis into crop species.

Analysis of rhesus monkey alphoid DNA suggests that it arose by tandem duplication of an ancestral monomer unit followed by independent variation within two adjacent monomers (one becoming more divergent than the other) before their amplification as a dimer unit to produce tandem arrays. The rhesus monkey alphoid DNA is a tandemly repeated, 343-bp dimer; the consensus dimer is over 98% homologous to the alphoid dimers reported for baboon and bonnet monkey, 81% homologous to the African green monkey alpha monomer, and less than 70% homologous to the more divergent human alphoid DNAs. The consensus dimer consists of two wings (I and II, 172 and 171 bp, respectively) that are only 70% homologous to each other, but share seven regions of exact homology. These same regions are highly conserved among the consensus sequences of the other cercopithecid alphoid DNAs. The three alpha-protein binding sites reported for African green monkey alpha DNA by F. Strauss and A. Varshavsky (Cell 37: 889-901, 1984) occur in wings I and II, but with one site altered in wing I. Two cloned dimer segments are 98% homologous to the consensus, each containing 8 single-base-pair differences within the 343-bp segment. Surprisingly, 37% of these differences occur in regions that are evolutionarily conserved in the alphoid consensus sequences, including the alpha-protein binding sites. Sequence variation in this highly repetitive DNA family may produce unique nucleosomal architectures for different members of an alphoid array. These unique architectures may modulate the evolution of these repetitive DNAs and may produce unique centromeric characteristics in primate chromosomes.