SciBet as a portable and fast single cell type identifier

• Published in: Nature communications Vol. 11; no. 1; p. 1818
• Main Authors: Li, Chenwei, Liu, Baolin, Kang, Boxi, Liu, Zedao, Liu, Yedan, Chen, Changya, Ren, Xianwen, Zhang, Zemin
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 14.04.2020; Nature Publishing Group; Nature Portfolio
• Subjects: 38; 38/39; 631/114/1305; 631/208/212/2019; Accuracy; Algorithms; Annotations; Cell size; Classification; Computer applications; Databases as Topic; Datasets; Feature selection; Gene expression; Gene Expression Regulation; Gene sequencing; Genomics; Humanities and Social Sciences; Humans; Hypotheses; Identification methods; Internet; Life sciences; Maximum likelihood estimation; multidisciplinary; Reproducibility of Results; Ribonucleic acid; RNA; Science; Science (multidisciplinary); Single-Cell Analysis
• ISSN: 2041-1723; 2041-1723
• DOI: 10.1038/s41467-020-15523-2

Fast, robust and technology-independent computational methods are needed for supervised cell type annotation of single-cell RNA sequencing data. We present SciBet, a supervised cell type identifier that accurately predicts cell identity for newly sequenced cells with order-of-magnitude speed advantage. We enable web client deployment of SciBet for rapid local computation without uploading local data to the server. Facing the exponential growth in the size of single cell RNA datasets, this user-friendly and cross-platform tool can be widely useful for single cell type identification. The increasing size of single cell sequencing data sets calls for scalable cell annotation methods. Here, the authors introduce SciBet, which uses a multinomial distribution model and maximum likelihood estimation for fast and accurate single cell identification.