This thesis shows how anonymous markers are used to assay genetic polymorphisms, and then describes the analysis of major biotic-stress related genomic motifs from diverse Musa accessions, before identifying Bacterial Artificial Chromosome (BAC) clones carrying biotic and abiotic stress-related DN MOTIFS. Microsatellite (SSR) and retroelement-related sequences are abundant and can be exploited as anonymous genetic markers in Musa. SSR primers designed within sequenced BAC clones identified polymorphisms in Musa. LTR-retrotransposon fragments, including Pseudoviridae (Tyl-copia-like) and Metaviridae (Ty3-gypsy-like) families were present as different families in BACs with no correspondence to particular Musa genomes, but are useful for understanding genome evolution. Primers designed from genomic and EST databases were exploited to characterize sequences containing Nucleotide Binding Sites (NBS) and Leucine-Rich Repeat (LRR) motifs (associated with disease resistance genes), and genes associated with tolerance to heat (HSP, 70HSP), salinity (STP) and drought (DRFP) stress. Neither NBS nor LRR sequences are conserved with respect to genome, indicating that R-genes are specific and exist independently. BAC libraries allow tagging of conserved domains of NBS, LRR, retroelement and SSR motifs, giving understanding of the genomic context and control of R-genes. Reliable characterisation of these domains in Musa is possible via PCR-based screening. The thesis gives abroad insight into genome organisation, evolution and diversity of major classes of R-genes, enabling progress towards gene discovery and exploitation for plant breeding.

M. Tech. (Department of Biosciences, Faculty of Applied and Computer Sciences) Vaal University of Technology / Most cultivated bananas and plantains (Musa spp. sect. Eumusa), originated from two wild diploid species, Musa acuminata Colla (AA) and Musa balbisiana Colla (BB), which contributed the A and B genomes, respectively. The two genomes confer different traits to a banana plant. Intra- and interspecific hybridization between the wild diploid species, somatic mutations and selection over many thousands of years has given rise to considerable genetic variability in cultivated bananas. Bananas are classified according to its genome composition and a number of morphological traits are used to identify the genomes of a plant. Morphological classification can be misleading since the morphology of plants can be affected by environmental factors. Molecular techniques to identify the genomes of banana have many advantages. The objective of this study was to develop a SCAR (sequence characterized amplified region) marker from a previously reported A genome-specific RAPD fragment that distinguish the A genome of banana from the B genome. This fragment designated OPA17600 was cloned, sequenced and used to design longer 20-mer SCAR primers. Verification of the SCAR primers for its fidelity to the A genome was carried out on a sample of 22 homo-and heterogenomic accessions representing landraces and hybrids of different ploidy and genome combinations. Out of six primers sets that were tested one set (SC3) produced a unique 600 bp in all the A genome containing banana accessions. However, these primers also amplified an 800 bp fragment in all the BB genotypes and some accessions containing the A and B genomes. While previous reports suggested that there was considerable differentiation between the A and B genomes, recent evidence points to the contrary. The presence of the A genome fragment in the B genome genotypes and accessions may be due to recombination between the two genomes, translocations and substitutions. The study concluded that the 600-bp SCAR sequence is conserved across the A genome in Musa and can be used to identify the A genome in banana classification and Musa breeding programmes.