A fast and efficient count-based matrix factorization method for detecting cell types from single-cell RNAseq data

• Published in: BMC systems biology Vol. 13; no. Suppl 2; p. 28
• Main Authors: Sun, Shiquan, Chen, Yabo, Liu, Yang, Shang, Xuequn
• Format: Journal Article
• Language: English
• Published: London BioMed Central 05.04.2019; BioMed Central Ltd
• Subjects: Algorithms; Binomial distribution; Biochemistry; Bioinformatics; Biomedical and Life Sciences; Brain; Brain stem; Cell cycle; Cells (Biology); Cellular and Medical Topics; Clustering; Computational Biology/Bioinformatics; Data analysis; Data processing; Datasets; Factorization; Gene expression; Gene sequencing; Genes; Genomes; Information management; Information processing; Life Sciences; Matrices (Mathematics); Methods; Pancreas; Physiological; Principal components analysis; Ribonucleic acid; RNA; RNA sequencing; Sequence Analysis, RNA; Simulation and Modeling; Single-Cell Analysis - methods; Source code; Stem cells; Systems Biology; Time Factors; Deep learning; Matrix factorization; Read count; Single-cell RNA sequencing
• ISSN: 1752-0509; 1752-0509
• DOI: 10.1186/s12918-019-0699-6

Background Single-cell RNA sequencing (scRNAseq) data always involves various unwanted variables, which would be able to mask the true signal to identify cell-types. More efficient way of dealing with this issue is to extract low dimension information from high dimensional gene expression data to represent cell-type structure. In the past two years, several powerful matrix factorization tools were developed for scRNAseq data, such as NMF, ZIFA, pCMF and ZINB-WaVE. But the existing approaches either are unable to directly model the raw count of scRNAseq data or are really time-consuming when handling a large number of cells (e.g. n >500). Results In this paper, we developed a fast and efficient count-based matrix factorization method (single-cell negative binomial matrix factorization, scNBMF) based on the TensorFlow framework to infer the low dimensional structure of cell types. To make our method scalable, we conducted a series of experiments on three public scRNAseq data sets, brain, embryonic stem, and pancreatic islet. The experimental results show that scNBMF is more powerful to detect cell types and 10 - 100 folds faster than the scRNAseq bespoke tools. Conclusions In this paper, we proposed a fast and efficient count-based matrix factorization method, scNBMF, which is more powerful for detecting cell type purposes. A series of experiments were performed on three public scRNAseq data sets. The results show that scNBMF is a more powerful tool in large-scale scRNAseq data analysis. scNBMF was implemented in R and Python, and the source code are freely available at https://github.com/sqsun .