Deniz Akdemir

The paper proposes a covariance-based method to combine partially overlapping relationship matrices from heterogeneous omics datasets to improve data harmonization and genomic prediction.

This study compares univariate and multivariate genomic prediction models for cassava traits, finding that multivariate models improve prediction accuracy in both single and multi-environment evaluations.

The paper proposes multi-objective optimized breeding strategies to achieve sustainable genetic gains across multiple traits while controlling co-ancestry in plant and animal breeding programs.

The paper introduces STPGA, a genetic algorithm for optimal subset selection, applied to genomic prediction and other statistical problems.