Needles

Needles models marker-by-environment interactions to improve genomic prediction of agronomic performance in future hybrid breeds.

Key Features:

• Marker-by-Environment Interaction: Models interactions between genetic markers and environmental conditions to capture variation in marker effects across environments.
• High-Performance Computing (HPC): Implements computations on a dedicated high-performance computing cluster to analyze large-scale genomic datasets.
• Sparse Matrix Formalism: Employs a sparse matrix formalism within its algebraic framework to reduce computational complexity by focusing on non-zero matrix elements.
• Distributed Computing Techniques: Distributes data and computations across multiple computing nodes to process large datasets concurrently.

Scientific Applications:

• Agronomic Performance Prediction: Predicts agronomic performance of hybrid breeds from genotypic marker data and marker-by-environment interactions.
• Quantitative Trait Loci (QTL) Discovery: Facilitates discovery of QTLs that contribute to performance under specific environmental conditions.

Methodology:

Integrates genotypic marker data with environmental interaction effects within a high-performance computing framework; manages large-scale trial datasets via distributed computing and sparse matrix formalism; analyzes simulated trial data to evaluate prediction accuracy, highlighting improved prediction of markers with high environmental interactions when more records per environment are available in the training dataset.