High throughput genotyping and phenotyping have revolutionized plant breeding, yet their integration remains limited in minor crops like lentils (Lens spp.). This study leverages phenotypic data derived from UAV-based imagery and manually measured data from multi-year trials on a diverse lentil population and integrates phenotypic and genotypic data for crop growth dissection and yield prediction. Using a Nested Association Mapping (NAM) population developed from CDC Redberry crossed with 32 genotypes, the study applies machine learning and crop growth modeling to identify key genomic regions controlling crop growth and development.

The whole thesis is structured into four research chapters. The first chapter focuses on developing the NAM population to improve the detection of minor-effect genes and enhance genetic resolution. The second chapter integrates traditional and UAV-based phenotyping along with crop growth modeling and machine learning approaches to analyze crop growth and predict seed yield. The third chapter employs Genome-Wide Association Studies (GWAS) to identify genomic regions associated with growth-related traits by incorporating all derived phenotypes. The final chapter investigates canopy collapse at later growth stages using UAV-derived phenotypes and crop growth modeling to identify QTLs associated with crop lodging or senescence.