Book excerpt: Precision medicine is a promising new approach to medicine that takes into account the individual differences in people's genetic makeup and lifestyle to identify specific treatment and prevention strategies for diseases. However, many human diseases are complex, and are driven by multiple layers of dysregulation at the cellular level, in addition to environmental factors. In recent years, the advances in high throughput technologies enable interrogation of biological systems at multiple levels, offering valuable types of data representing various aspects of cellular systems. These data types include sequences and structures of genes, RNAs, proteins, quantitative measurements on the abundance of these molecules under different conditions, and the interactions among these molecules. However, these data are noisy, incomplete, high-dimensional, highly heterogeneous, and often provide static representations of a complex and dynamic system. In this thesis, we develop computational methods to make use of these useful, yet limited sources of biological data, with a view to gaining insights on the molecular mechanisms of complex diseases. In particular, we develop novel algorithms to integrate genomic (genome-wide association studies), transcriptomic (expression-quantitative trait locus interactions), proteomic (protein expression screened via mass spectrometry), phospho-proteomic (large scale data on the phosphorylation of signaling proteins screened via mass spectrometry), and interactomic (protein interaction networks, pathway databases) datasets. Using these integrative algorithms, we develop computational tools for the identification of disease-associated protein subnetworks, risk assessment for complex diseases, and prediction of kinase-substrate associations in specific biological contexts.