SAME-clustering: Single-cell RNA-seq Aggregated clustering via Mixture model Ensemble

SAME-clustering is a flexible, accurate and robust method for clustering scRNA-seq data. SAME-clustering takes as input, results from multiple clustering methods to build one consensus clustering. Specifically, in our current implementation, single cells are first clustered using five state-of-the-art methods, SC3, CIDR, Seurat, t-SNE + k-means, and SIMLR. Of the five sets of solutions, we choose a maximally diverse subset of four according to variation in pairwise Adjusted Rand Index (ARI). The four individual solutions are then ensembled by solving a Multinomial Mixture Model via the EM algortihm. For details, please refer to our tutorial.

Comments and suggestions are welcome, please e-mail Ruth Huh at rhuh@live.unc.edu or Yun Li at yunli@med.unc.edu.