Variable selection in heterogeneous datasets: A truncated-rank sparse linear mixed model with applications to genome-wide association studies

Keyphrases

• Heterogeneous Data 100%
• Genome-wide Association Study 100%
• Population Structure 100%
• Linear Mixed Model 100%
• Biological Datasets 50%
• Unified Framework 50%
• Low-rank 50%
• Sample Structure 50%
• Multi-population 50%
• Covariance Structure 50%
• Sparse Parameters 50%
• Variable Selection Methods 50%
• Least Absolute Shrinkage and Selection Operator (LASSO) 50%
• Medical Dataset 50%
• False Discovery 50%
• Genomic Datasets 50%
• Non-IID 50%
• Association Studies in Genetics 50%

Computer Science

• Feature Selection 100%
• Wide Association Study 100%
• Unified Framework 20%
• False Discovery 20%
• Renewed Interest 20%
• Underlying Population 20%
• Selection Method 20%

Mathematics

• Linear Mixed Model 100%
• Subpopulation 50%
• Renewed Interest 50%
• Covariance Structure 50%
• Classical Variable 50%