The most efficient, effective, environmentally safe and economical means to control downy mildew of pearl millet is the use of resistant cultivars. Earlier studies on this host-pathogen interaction have shown that resistance is polygenically controlled. The parental lines IP 18293 (downy mildew resistant, d2 dwarf, purple foliaged) and Tift 238D1 (downy mildew susceptible, green foliaged, d1 dwarf) were used for this new mapping population. Replicated screening of F2:F4 segregating progenies from the (IP 18293 x Tift 238D1)-based mapping population, against downy mildew pathogen populations from India (Patancheru, Jalna, Jodhpur, Jamnagar, Durgapura and New Delhi) were done at ICRISAT and against African pathogen populations (Niger and Mali) at Bangor, UK. The segregating F2:F4 progenies showed continuous disease reaction inheritance patterns that varied across the pathogen populations. Depending on the pathogen population, resistance appeared to be controlled by monogenic, digenic or trigenic Mendelian ratios. From the observed segregation patterns, it was shown that at least three genes were controlling resistance to this range of pathogen populations.

Cluster analysis and Pearson rank correlation values for F2:F4 progeny disease incidence revealed relationships between the diverse Indian and African downy mildew pathogen populations. The northern Indian pathogen populations (Duragapura and New Delhi) and the African pathogen populations (Niger and Mali) were shown to be more virulent than those from southern and central India (Patancheru, Jalna, Jamnagar and Jodhpur), at least with regard to the host-plant resistance segregating in this mapping population.

Detailed genetic linkage maps in plants are very useful tools for studying genome structure and evolution, identifying introgression between genomes, and localizing genes of interest. RFLP markers have simple genetic segregation patterns and are potentially unlimited in number. Detailed RFLP linkage maps have been constructed for several important crops such as maize, tomato, lettuce, potato and rice. The first RFLP-based genetic linkage map of pearl millet was constructed by Liu et al. (1994). In the present study an (IP 18293 x Tift 238D1)-based F2:F3 population was used for map construction and its F2:F4 segregating progenies were phenotyped for their resistance behaviour against eight downy mildew pathogen populations. The amount of marker polymorphism exhibited in this pearl millet mapping population was 40%, as high as that of maize and other outbreeding crop species. From the 220 probe-enzyme combinations assessed in parental screening, 33 RFLP probe-enzyme combinations were selected for linkage map construction. Using these 33 selected RFLP loci a skeleton map with a length of 561.8 cM was produced for the (IP 18293 x Tift 238D1)-based mapping population using Mapmaker/Exp. Most of the selected markers segregated as expected with the Mendelian segregation ratio of 1:2:1. However, 33% of marker the loci exhibited distorted segregation. This was mostly because of the excess of heterozygotes and alleles from male parent Tift 238D1. Marker orders in all linkage groups were the same as base map of Liu et al. (1994). Along with this base map, the linkage map for this population was compared with all the previously constructed pearl millet maps.

Although the use of most conventional morphological markers is not usually practicable in breeding programs, those that are available should not be ignored. In this present study, marker loci controlling dwarfness (d1 and d2) and purple foliage colour (P) were used along with the RFLP markers. The d1 dwarfing gene has been placed on the top of the linkage group 1 and the d2 dwarfing gene on linkage group 4. The purple colour locus P was also placed on linkage group 4 using this newly constructed (IP 18293 x Tift 238D1)-based pearl millet mapping population.

Molecular markers are rapidly being adopted by crop improvement researchers globally as an effective and appropriate tool for basic and applied studies of biological components in agricultural systems. Use of markers in applied breeding programs can range from facilitating appropriate choice of parents for crosses, to mapping/tagging of gene blocks associated with economically important traits. Reports on linkage between quantitative trait loci effects and marker genotypes have been available in the literature quite a long time. Molecular markers tightly linked to different disease resistance genes have potential importance in facilitating selection procedures particularly for pyramiding two or more different resistance genes with the intension of producing a more durable and broad-spectrum resistance. In this study host-plant resistance QTLs were identified from parental line IP 18239 for six Indian and two African downy mildew pathogen populations. For QTL mapping, the interval-mapping method implemented in Mapmaker/QTL and the composite interval mapping method from QTL Cartographer were used. A total of seven different resistance QTLs were identified from screens of the mapping population progenies against these eight different pathogen populations. Among these, a common resistance QTL was identified on linkage group 2, which was effective against four Indian pathogen populations (Patancheru, Jodhpur, Jalna and Jamnagar). Such disease resistances are expected to be durable for a number of years. Likewise a common resistance QTL was identified on linkage group 4 for the African pathogen populations from Sadore, Niger and Bamako, Mali. These resistance QTLs are of considerable interest because the durability of resistance is of major concern for plant breeders. Several pathogen populations specific resistant to downy mildew QTLs were also identified, but these are not expected to contribute to durable resistance unless deployed as components of uniform or segregating resistance gene pyramids.

QTL mapping and DNA markers also provide insights into facets of quantitative inheritance patterns. In this present mapping population all identified resistance QTLs were from the resistant parent IP 18293, and for these an over-dominant inheritance pattern was most commonly observed. These identified resistance QTLs can now be transferred to genetic backgrounds of elite hybrid parental lines through marker-assisted selection breeding programs. Flanking markers of the identified QTLs can facilitate selection of resistant progenies during this backcrossing process, whereas other marker loci can be used in reducing the length of the donor segments carried along with the introgressed genes and/or selecting for recovery of recurrent parent alleles on non-carrier chromosomes. Marker-assisted selection can also be used to a pyramid several resistant genes into a single male-sterile line (and its maintainer) or pollinator line.