Finding cancer subtypes in microarray data using random projections

One of the benefits of profiling cancer samples using microarrays is the generation of molecular fingerprints that will define subtypes of disease. Such subgroups have typically been found in microarray data using hierarchical clustering. A major problem in the interpretation of the output is determining the number of clusters. We approach the problem of determining disease subtypes using mixture models. A novel estimation procedure for the parameters in the mixture model is developed based on a combination of random projections and the expectation–maximization algorithm. Because the approach is probabilistic, our method provides a measure for the number of true clusters in a given dataset. We illustrate our proposal with applications to both simulated and real microarray data.