**13. Predictive breeding for resistance to** *L. maculans* **using molecular markers**

Success of new disease resistance genes relies heavily on the successful transfer of target genomic regions from donor sources and the development of rigorous selection methods. Molecular markers have been used to improve the effectiveness and efficiency of selection strategies in predictive breeding in several agricultural crops. However, the development of molecular markers in *B. napus* and their application in breeding is a challenging exercise due to the large genome size, amphidipliod (4X) nature, open-pollination and lower research funding as compared to other key crops such as wheat, barley, maize and soybean. The *B. napus* genome is highly complex and homologous recombination plays a major role in chromosome rearrangements such as duplications and reciprocal translocations. These arrangements further add to the complexity of molecular analysis and interpretation. *B. napus* chromosomes C6 and A7, which harbours *Rlm1*, *Rlm3, Rlm4, Rlm7* and *Rlm9* genes for resistance, produced a reciprocal translocation in some cultivars such as in Westar, Marnoo, Monty and Maluka [185, 186] which makes analysis of resistance genes difficult [142].

In most of the breeding programs, selection for blackleg is conducted once a year during the growing season, hampering selection efficiency. Several studies suggest a significant correla‐ tion between cotyledon test and canker lesion scores. Therefore, cotyledon tests can be used for selection for resistance to *L. maculans*. However, in many developed countries, it is costly and laborious to perform, particularly as compared to molecular marker analysis, when several tests need to be carried to screen large populations. Furthermore, analysis of different blackleg resistance genes in a canola breeding program using a differential set of *L. maculans* isolates at various stages of the breeding cycle is a very slow process [39]. Interpretation of *R*-gene content using a differential set of control *B. napus* varieties, especially of Australian origin, is a challenging exercise, as majority of cultivars used are heterozygous and/or heterogeneous [32, 41]. In addition, phenotypic tests are dependent upon the growing environment (microclimate conditions and other factors such as powdery and downy mildew), which can complicate scoring of inoculated seedlings. Molecular markers generally out-perform conventional seedling assays, in both efficiency and reliability. It is also possible to identify haplotypes using molecular markers and then validate trait-marker associations, in conjunction with compre‐ hensive phenotyping and conventional allelism tests.

The published literature suggests that little effort has been made to evaluate the allelic relationship among the known genes from different sources, to test stability of majority of QTL or qualitative genes identified over diverse growing environments, or to test their usefulness in achieving long term durable control of the disease. Table 1 also suggests that majority of markers are not very closely linked (<1cM) with resistance loci. Diagnostic or perfect markers for resistance genes are required for routine MAS and will assist allele enrichment strategies in breeding programs, although this is not always possible, even if the complete gene is cloned and characterised for its functionality [187]. The linkage between molecular markers and *Xbn204* flanking the *RlmSkipton* locus was verified in an F2 population derived from Skipton/Ag-Spectrum [32]. The results showed that SSR markers linked to *RlmSkipton* are suitable for enrichment of favourable alleles for blackleg resistance in breeding programs. A separate study [82] validated the map location of *Rlm1* in the DH population derived from Maxol/Westar with SSR and DArT markers. Previously, *Rlm1* and *Rlm3* genes were mapped on chromosomes A7 in the Maxol (resistant to blackleg)/S006 (susceptible to blackleg) utilising RAPD markers and with single spore isolates with known *Avr* genotypes in the *B. napus* European cultivars, Columbus and Maxol [41, 71, 74]. RAPD markers are not amenable for high throughput marker analysis, as they are assayed on low-throughput agarose or polya‐ crylamide gel systems. Validation of a large array of genes for blackleg resistance in diverse segregating populations representing *B. napus* germplasm is a challenging exercise. However, an association mapping approach can be employed to test trait-marker associations in a large set of germplasm as demonstrated recently [137, 142].
