Abstract
PURPOSE: Septoria tritici blotch, caused by the ascomycete fungus Zymoseptoria tritici, is one of the major threats to global food security, significantly reducing wheat yields. Current control strategies rely heavily on fungicide applications, which can accelerate fungal adaptation and resistance, highlighting the urgent need for sustainable and durable disease management approaches. METHODS: We performed genome-wide association studies in wheat and fungal populations to identify novel marker-trait associations linked to virulence within the pathosystem. Additionally, we integrated marker information from both the host and pathogen to evaluate whether a dual-source approach could improve genomic prediction of wheat resistance and reveal specific host-pathogen interactions. RESULTS: We identified five novel marker-trait associations potentially involved in pathogen recognition across six wheat chromosomes and two overlapping known QTL regions. On the pathogen side, we identified 29 candidate genes potentially associated with fungal virulence, including an effector-like protein. The integrated host-pathogen genomic selection models improved predictive accuracy by capturing both wheat genotype and pathogen variation, although host genetics explained most of the variation. CONCLUSION: Conventional single-genome models lacked power, underscoring the value of dual-genome approaches in complex pathosystems. These findings support modeling genotype-by-genotype interactions for breeding resistant cultivars while concurrently monitoring pathogen evolution.