Skip to main content
. Author manuscript; available in PMC: 2017 Jan 2.
Published in final edited form as: Nat Rev Genet. 2015 May 7;16(6):321–332. doi: 10.1038/nrg3920

Figure 5. Three ways to accommodate heterogeneous data in machine learning.

Figure 5

The task of predicting gene function labels requires methods that take as input gene expression data, protein sequences, protein-protein interaction networks, etc. These diverse data types can be encoded into fixed-length features, represented using pairwise similarities (kernels), or directly accommodated by a probability model.