Bogdan Pasaniuc, Alkes Price, Peter Kraft and colleagues report a new integrative statistical method for the fine mapping of associations assisted by functional annotation, which is useful in prioritizing variants to include in following functional studies (PLoS Genet. 10, e1004722, 2014). Their method, PAINTOR (Probabilistic Annotation INTegratOR), provides a framework for incorporating functional annotations with association statistics to assign a probability of causality for SNPs at a previously associated locus. The authors used an empirical Bayes prior to integrate functional annotation data and maximum-likelihood estimation to simultaneously estimate model parameters over all fine-mapping loci. Importantly, their method allows for multiple causal variants at any locus. The authors demonstrate in simulations that using posterior probabilities to prioritize variants provides greater accuracy than other fine-mapping methods in identifying causal variants, in part because the method allows for multiple causal variants. They apply their approach to a large meta-analysis for 4 lipid traits in combination with 450 cell type–specific annotations. They demonstrate that their approach improves the prioritization of causal variants, reducing the 90% confidence set from 17.5 to 13.5 SNPs per locus. The software is publicly available at http://bogdan.bioinformatics.ucla.edu/software/paintor.