To address deficiencies of condition-by-condition analyses, several groups have developed methods for joint analysis of multiple conditions2–3,5–15. The simplest methods build on traditional meta-analysis methodology8,9, and assume that non-zero effects are shared across all conditions. Other methods are more flexible, allowing for condition-specific effects, for sharing of effects among subsets of conditions, and for heterogeneity in the shared effects5,6,12. The most flexible methods adapt themselves to each dataset by learning patterns of sharing using a hierarchical model5.
