One of the primary tasks facing biologists today is to integrate the different views of molecular biology that are provided by various types of experimental data. In yeast, for example, for a given gene we typically know the protein it encodes, that protein’s similarity to other proteins, the mRNA expression levels associated with the given gene under hundreds of experimental conditions, the occurrences of known or inferred transcription factor binding sites in the upstream region of that gene, and the identities of many of the proteins that interact with the given gene’s protein product. Each of these distinct data types provides one view of the molecular machinery of the cell. Kernel methods allow us to represent these heterogeneous data types in a normal form, and to use kernel algebra to reason about more than one type of data simultaneously. Consequently, multi-kernel learning methods have been applied to a variety of biology applications. In this talk, I will describe several of these applications, outline the lessons we have learned from applying multi-kernel learning methods to real data, and suggest several avenues for future research in this area.