Genome-wide association and genomic prediction of resistance to maize lethal necrosis disease in tropical maize germplasm

Genome-wide association analysis in tropical and subtropical maize germplasm revealed that MLND resistance is influenced by multiple genomic regions with small to medium effects. The maize lethal necrosis disease (MLND) caused by synergistic interaction of Maize chlorotic mottle virus and Sugarcane mosaic virus, and has emerged as a serious threat to maize production in eastern Africa since 2011. Our objective was to gain insights into the genetic architecture underlying the resistance to MLND by genome-wide association study (GWAS) and genomic selection. We used two association mapping (AM) panels comprising a total of 615 diverse tropical/subtropical maize inbred lines. All the lines were evaluated against MLND under artificial inoculation. Both the panels were genotyped using genotyping-by-sequencing. Phenotypic variation for MLND resistance was significant and heritability was moderately high in both the panels. Few promising lines with high resistance to MLND were identified to be used as potential donors. GWAS revealed 24 SNPs that were significantly associated (P < 3 × 10(-5)) with MLND resistance. These SNPs are located within or adjacent to 20 putative candidate genes that are associated with plant disease resistance. Ridge regression best linear unbiased prediction with five-fold cross-validation revealed higher prediction accuracy for IMAS-AM panel (0.56) over DTMA-AM (0.36) panel. The prediction accuracy for both within and across panels is promising; inclusion of MLND resistance associated SNPs into the prediction model further improved the accuracy. Overall, the study revealed that resistance to MLND is controlled by multiple loci with small to medium effects and the SNPs identified by GWAS can be used as potential candidates in MLND resistance breeding program.