Figure 1.

This figure illustrates how different data types can be coupled to each other Example (a): Meta-analysis or horizontal integrative analysis involves data collection under different conditions resulting in two datasets that share the same features (e.g., only metabolomic features) but different samples. These observations can be combined into one data matrix after meta-analysis. Example (b): In heterogeneous or vertical integrative analysis data are acquired from samples profiled under the same conditions, but do not share the same features e.g., genomic features vs metabolomic features. Strategies that can be used for these types of integrative analysis are depicted in Figure 3.