This book explains and demonstrates with real and simulated examples how whole-genome information can be used for predicting complex traits, with applications in animal, human, and plant genetics. After giving a brief introduction, the book covers linear models and dimensionality, plus regularized regressions. It then progresses to the genomic best linear unbiased predictor, the Bayesian alphabet, reproducing Kernel Hiblert spaces regressions, penalized neural networks, and re-sampling methods. Lastly, it covers whole genome regression and population stratification.