Presentation Title: Single-cell RNA-sequencing analysis of human inner ear organoids using AutoClustR: A computational tool for unbiased cell type discovery
Author Name(s): Daniel R. Romano, Alex Solivais, Jing Nie, Eri Hashino
Abstract: Sensorineural hearing loss is the most common form of permanent hearing loss, which is itself among the most prevalent sensory impairments. While hearing aids and cochlear implants provide great benefit to many hearing-impaired individuals, they are not curative. More precise approaches will undoubtedly require a deeper understanding of developmental and pathological processes in the inner ear. Droplet-based single-cell RNA sequencing (scRNA-seq) has recently emerged as a powerful tool for dissecting these processes. Unsupervised clustering is a crucial step in scRNA-seq analysis, as cell clusters are presumed to correspond with distinct cell types, subtypes, and states. However, current clustering algorithms are driven by cluster number estimates, which are not readily available for potentially fruitful applications of scRNA-seq, such as the discovery of cell types and the validation, optimization, and comparison of stem cell differentiation protocols. To address this problem, we developed AutoClustR, an R-based computational tool for unbiased clustering of scRNA-seq data. Using seventeen datasets, we show that AutoClustR outperforms two similarly focused tools. Next, we used AutoClustR to reveal an unprecedented – and previously unappreciated – cellular diversity within stem cell-derived inner ear organoids. We envisage AutoClustR as invaluable to the ultimate development of cell-based and gene therapies for sensorineural hearing loss.