Cluster Aware Deep Dictionary Learning for Single Cell Analysis

Abstract

The importance of clustering the single-cell RNA sequence is well known. Traditional clustering techniques (GiniClust, Seurat, etc.) have mostly been used to address this problem. This is the first work that develops a deep dictionary learning-based solution for the same. Our work builds on the framework of deep dictionary learning. We make the framework clustering friendly by incorporating a cluster-aware loss (K-means and sparse subspace) into the learning problem. Comparison with tailored clustering techniques for single-cell RNA and with generic deep learning-based clustering techniques shows the promise of our approach.