MAZE explored the potential of novel experimental populations derived from maize landraces. A diversity atlas aggregating genetic material, data and results from the first phase will be established and extended by expression data and germplasm segregating for drought and heat tolerance (WP1). We will study novel alleles and functionally characterize candidate genes and pathways involved in abiotic stress tolerance with forward and reverse genetic approaches (WP2). In the context of whole-genome prediction, we will investigate the relevance of integrating G×G interactions, airborne imaging, endophenotypes, and climate data for prediction model building (WP3). By devising genome-based rapid cycling breeding schemes for maize genetic resources through simulations and large-scale selection experiments, we will assess the realized and expected selection gain for multiple traits and quantify the genetic load present in maize landraces (WP4). The project will yield insights into genes contributing to abiotic stress tolerance, their function, interplay and trade-offs.