Fast and interpretable consensus clustering via minipatch learning

Consensus clustering has been widely used in bioinformatics and other applications to improve the accuracy, stability and reliability of clustering results. This approach ensembles cluster co-occurrences from multiple clustering runs on subsampled observations. For application to large-scale bioinformatics data, such as to discover cell types from single-cell sequencing data, for example, consensus clustering has two significant drawbacks: (i) computational inefficiency due to repeatedly applying clustering algorithms, and (ii) lack of interpretability into the important features for differentiating clusters. In this paper, we address these two challenges by developing IMPACC: Interpretable MiniPatch Adaptive Consensus Clustering. Our approach adopts three major innovations. We ensemble cluster co-occurrences from tiny subsets of both observations and features, termed minipatches, thus dramatically reducing computation time. Additionally, we develop adaptive sampling schemes for observations, which result in both improved reliability and computational savings, as well as adaptive sampling schemes of features, which lead to interpretable solutions by quickly learning the most relevant features that differentiate clusters. We study our approach on synthetic data and a variety of real large-scale bioinformatics data sets; results show that our approach not only yields more accurate and interpretable cluster solutions, but it also substantially improves computational efficiency compared to standard consensus clustering approaches.Author summary: Clustering seeks to discover groups in big data with wide applications across scientific domains, especially in bioinformatics. However, for huge and sparse data sets common with genomic sequencing technologies, clustering methods can suffer from unreliable results, lack of interpretability in terms of feature importance, and heavy computational costs. To solve these challenges, we propose an extension of consensus clustering that leverages minipatch learning, an ensemble learning framework with learners trained on tiny subsets of observations and features. With adaptive sampling frameworks on both features and observations, our method is able to achieve higher clustering accuracy and reliability, as well as simultaneously identify scientifically important features that distinguish the clusters. In addition, we offer major computational improvements, with dramatically faster speed than our competitors. Our method is general and widely applicable to data sets from any field, and especially can offer superior performance when dealing with complex sparse and high dimensional data found in bioinformatics.