As datasets grow, computational efficiency becomes a statistical problem. This part covers network analysis, clustering, and machine learning (Random Forests, SVMs) applied to genomic data.
The is the definitive bridge between the mathematician’s theorem and the biologist’s hypothesis. For a researcher staring at a spreadsheet of 20 million genetic variants and asking, "What does this mean?", Balding provides the vocabulary, the methods, and the statistical caution required to find a valid answer. Balding D. Handbook of Statistical Genomics 2019
Gene expression, epigenetics, metabolomics, and microbiome analysis. Application Areas: For a researcher staring at a spreadsheet of
In 2009, a GWAS of 5,000 individuals was heroic. In 2019, studies with 500,000 individuals are routine. The new edition includes extensive coverage of linear mixed models (LMMs) , which account for cryptic relatedness and population stratification at scale (e.g., the BOLT-LMM and SAIGE algorithms). In 2019, studies with 500,000 individuals are routine