Abstract
The dramatic increase in amount and size of single-cell RNA sequencing data calls for more efficient and scalable dimensional reduction and visualization tools. Here, we design a GPU-accelerated method, NeuralEE, which aggregates the advantages of elastic embedding and neural network. We show that NeuralEE is both scalable and generalizable in dimensional reduction and visualization of large-scale scRNA-seq data. In addition, the GPU-based implementation of NeuralEE makes it applicable to limited computational resources while maintains high performance, as it takes only half an hour to visualize 1.3 million mice brain cells, and NeuralEE has generalizability for integrating newly generated data.
Highlights
Dimensionality reduction is one of the basic steps in machine learning algorithms and big-data analyses, especially in the analysis of high-throughput single cell RNA sequencing data. scRNA-seq enables us to simultaneously profile thousands of genetic markers at singlecell resolution, which makes it an ideal tool to study the cell-cell heterogeneity in developmental biology, oncology, and immunology
NeuralEE, NeuralEE-SO, and EE exhibited the best performance with NeuralEE and EE ties at the top
We develop NeuralEE, a GPU-accelerated dimensionality reduction method for visualization of large-scale scRNA-seq data
Summary
Dimensionality reduction is one of the basic steps in machine learning algorithms and big-data analyses, especially in the analysis of high-throughput single cell RNA sequencing data (scRNA-seq data). scRNA-seq enables us to simultaneously profile thousands of genetic markers at singlecell resolution, which makes it an ideal tool to study the cell-cell heterogeneity in developmental biology, oncology, and immunology. Dimensionality reduction is one of the basic steps in machine learning algorithms and big-data analyses, especially in the analysis of high-throughput single cell RNA sequencing data (scRNA-seq data). Visualization of scRNA-seq data in a manageable dimension often plays as a pivotal first step prior to other downstream analyses such as cell type identification or cell developmental trajectory reconstruction. Among the numerous dimensionality reduction and visualization methods, t-distributed stochastic neighbor embedding (t-SNE) (van der Maaten and Hinton, 2008) is most widely used in the single-cell community to visualize data structures. As an extension of stochastic neighbor embedding (SNE), elastic embedding (EE) algorithm penalizes, placing far apart latent points from similar data points and placing close together latent points from dissimilar data points (Carreira-Perpinán, 2010), thereby preserving the intrinsic data structure both locally and globally (Hie et al, 2020). EE has been recently proved to be well-performed in visualization and in reconstruction of the embedded structure of the cell developmental process (An et al, 2019; Chen et al, 2019)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.