The role of genomics in personalized medicine continues to undergo profound changes, in step with dramatic technological advances. Ability to sequence the entire human genome with relative ease raises expectations that we can use an individual's complete genomic blueprint to understand disease risk and predicting therapy outcomes, thereby, optimizing drug therapy. Yet, doubts persist as to what extent genetic/genomic factors influence disease and treatment outcomes or whether robust predictive biomarker tests can be developed. Encompassing more than just DNA sequences, the definition of genomics now often is taken to include transcriptomics, proteomics, metabolomics, and epigenomics, with integration of genomic and environmental factors, in an area referred to systems biology. While we can learn much about a cell's innermost workings, summation of these diverse areas is far from enabling the prediction of therapeutic outcomes. Typically, only a handful of specific biomarkers, genetic or otherwise, are 'actionable', i.e., they can be used to guide therapy. I will focus on pharmacogenetic biomarkers, highlighting current successes but also the main challenges that remain in optimizing individualized therapy.