Large-scale Single-cell RNA-seq Data Research Articles

CIForm as a Transformer-based model for cell-type annotation of large-scale single-cell RNA-seq data.

Single-cell omics technologies have made it possible to analyze the individual cells within a biological sample, providing a more detailed understanding of biological systems. Accurately determining the cell type of each cell is a crucial goal in single-cell RNA-seq (scRNA-seq) analysis. Apart from overcoming the batch effects arising from various factors, single-cell annotation methods also face the challenge of effectively processing large-scale datasets. With the availability of an increase in the scRNA-seq datasets, integrating multiple datasets and addressing batch effects originating from diverse sources are also challenges in cell-type annotation. In this work, to overcome the challenges, we developed a supervised method called CIForm based on the Transformer for cell-type annotation of large-scale scRNA-seq data. To assess the effectiveness and robustness of CIForm, we have compared it with some leading tools on benchmark datasets. Through the systematic comparisons under various cell-type annotation scenarios, we exhibit that the effectiveness of CIForm is particularly pronounced in cell-type annotation. The source code and data are available at https://github.com/zhanglab-wbgcas/CIForm.

SUSCC: Secondary Construction of Feature Space based on UMAP for Rapid and Accurate Clustering Large-scale Single Cell RNA-seq Data.

Clustering is a common method to identify cell types in single cell analysis, but the increasing size of scRNA-seq datasets brings challenges to single cell clustering. Therefore, it is an urgent need to design a faster and more accurate clustering method for large-scale scRNA-seq data. In this paper, we proposed a new method for single cell clustering. First, a count matrix is constructed through normalization and gene filtration. Second, the raw data of gene expression matrix are projected to feature space constructed by secondary construction of feature space based on UMAP (Uniform Manifold Approximation and Projection). Third, the low-dimensional matrix on the feature space is randomly divided into two sub-matrices according to a certain proportion for clustering and classifying, respectively. Finally, one subset is clustered by k-means algorithm and then the other subset is classified by k-nearest neighbor algorithm based on clustering results. Experimental results show that our method can cluster the scRNA-seq datasets effectively.

GRNdb: decoding the gene regulatory networks in diverse human and mouse conditions.

Gene regulatory networks (GRNs) formed by transcription factors (TFs) and their downstream target genes play essential roles in gene expression regulation. Moreover, GRNs can be dynamic changing across different conditions, which are crucial for understanding the underlying mechanisms of disease pathogenesis. However, no existing database provides comprehensive GRN information for various human and mouse normal tissues and diseases at the single-cell level. Based on the known TF-target relationships and the large-scale single-cell RNA-seq data collected from public databases as well as the bulk data of The Cancer Genome Atlas and the Genotype-Tissue Expression project, we systematically predicted the GRNs of 184 different physiological and pathological conditions of human and mouse involving >633 000 cells and >27 700 bulk samples. We further developed GRNdb, a freely accessible and user-friendly database (http://www.grndb.com/) for searching, comparing, browsing, visualizing, and downloading the predicted information of 77 746 GRNs, 19 687 841 TF-target pairs, and related binding motifs at single-cell/bulk resolution. GRNdb also allows users to explore the gene expression profile, correlations, and the associations between expression levels and the patient survival of diverse cancers. Overall, GRNdb provides a valuable and timely resource to the scientific community to elucidate the functions and mechanisms of gene expression regulation in various conditions.

ScAIDE: clustering of large-scale single-cell RNA-seq data reveals putative and rare cell types.

Recent advancements in both single-cell RNA-sequencing technology and computational resources facilitate the study of cell types on global populations. Up to millions of cells can now be sequenced in one experiment; thus, accurate and efficient computational methods are needed to provide clustering and post-analysis of assigning putative and rare cell types. Here, we present a novel unsupervised deep learning clustering framework that is robust and highly scalable. To overcome the high level of noise, scAIDE first incorporates an autoencoder-imputation network with a distance-preserved embedding network (AIDE) to learn a good representation of data, and then applies a random projection hashing based k-means algorithm to accommodate the detection of rare cell types. We analyzed a 1.3 million neural cell dataset within 30 min, obtaining 64 clusters which were mapped to 19 putative cell types. In particular, we further identified three different neural stem cell developmental trajectories in these clusters. We also classified two subpopulations of malignant cells in a small glioblastoma dataset using scAIDE. We anticipate that scAIDE would provide a more in-depth understanding of cell development and diseases.

Latent cellular analysis robustly reveals subtle diversity in large-scale single-cell RNA-seq data.

Single-cell RNA sequencing (scRNA-seq) is a powerful tool for characterizing the cell-to-cell variation and cellular dynamics in populations which appear homogeneous otherwise in basic and translational biological research. However, significant challenges arise in the analysis of scRNA-seq data, including the low signal-to-noise ratio with high data sparsity, potential batch effects, scalability problems when hundreds of thousands of cells are to be analyzed among others. The inherent complexities of scRNA-seq data and dynamic nature of cellular processes lead to suboptimal performance of many currently available algorithms, even for basic tasks such as identifying biologically meaningful heterogeneous subpopulations. In this study, we developed the Latent Cellular Analysis (LCA), a machine learning–based analytical pipeline that combines cosine-similarity measurement by latent cellular states with a graph-based clustering algorithm. LCA provides heuristic solutions for population number inference, dimension reduction, feature selection, and control of technical variations without explicit gene filtering. We show that LCA is robust, accurate, and powerful by comparison with multiple state-of-the-art computational methods when applied to large-scale real and simulated scRNA-seq data. Importantly, the ability of LCA to learn from representative subsets of the data provides scalability, thereby addressing a significant challenge posed by growing sample sizes in scRNA-seq data analysis.

Single Cell Explorer, collaboration-driven tools to leverage large-scale single cell RNA-seq data

BackgroundSingle cell transcriptome sequencing has become an increasingly valuable technology for dissecting complex biology at a resolution impossible with bulk sequencing. However, the gap between the technical expertise required to effectively work with the resultant high dimensional data and the biological expertise required to interpret the results in their biological context remains incompletely addressed by the currently available tools.ResultsSingle Cell Explorer is a Python-based web server application we developed to enable computational and experimental scientists to iteratively and collaboratively annotate cell expression phenotypes within a user-friendly and visually appealing platform. These annotations can be modified and shared by multiple users to allow easy collaboration between computational scientists and experimental biologists. Data processing and analytic workflows can be integrated into the system using Jupyter notebooks. The application enables powerful yet accessible features such as the identification of differential gene expression patterns for user-defined cell populations and convenient annotation of cell types using marker genes or differential gene expression patterns. Users are able to produce plots without needing Python or R coding skills. As such, by making single cell RNA-seq data sharing and querying more user-friendly, the software promotes deeper understanding and innovation by research teams applying single cell transcriptomic approaches.ConclusionsSingle cell explorer is a freely-available single cell transcriptomic analysis tool that enables computational and experimental biologists to collaboratively explore, annotate, and share results in a flexible software environment and a centralized database server that supports data portal functionality.

SSCC: A Novel Computational Framework for Rapid and Accurate Clustering Large-scale Single Cell RNA-seq Data

Clustering is a prevalent analytical means to analyze single cell RNA sequencing (scRNA-seq) data but the rapidly expanding data volume can make this process computationally challenging. New methods for both accurate and efficient clustering are of pressing need. Here we proposed Spearman subsampling-clustering-classification (SSCC), a new clustering framework based on random projection and feature construction, for large-scale scRNA-seq data. SSCC greatly improves clustering accuracy, robustness, and computational efficacy for various state-of-the-art algorithms benchmarked on multiple real datasets. On a dataset with 68,578 human blood cells, SSCC achieved 20% improvement for clustering accuracy and 50-fold acceleration, but only consumed 66% memory usage, compared to the widelyused software package SC3. Compared to k-means, the accuracy improvement of SSCC can reach 3-fold. An R implementation of SSCC is available at https://github.com/Japrin/sscClust.

Abstract 5297: LCA: A robust and scalable algorithm to reveal subtle diversity in large-scale single-cell RNA-Seq data

Abstract Single-cell RNA sequencing (scRNA-seq) emerges as a powerful tool to characterize cell-to-cell variation and dynamics in a seemingly homogenous population. Efficient and affordable, scRNA-seq is gaining in popularity in both basic and translational biological research areas. However, significant challenges arise in the analysis of scRNA-seq data, including low signal-to-noise ratio with high data sparsity, rising scalability hurdles with hundreds of thousands of cells, and more. Due to inherent complexities in scRNA-seq data, the performance of currently available algorithms may not always be optimal even for fundamental tasks such as identifying heterogeneous subpopulations in the data. In this study, we developed Latent Cellular Analysis (LCA), a machine learning based analytical pipeline that combines similarity measurement by latent cellular states with a graph-based clustering algorithm. LCA features a dual-space model search for both the optimal number of subpopulations and the informative cellular states distinguishing them. LCA provides heuristic solutions for population number inference, dimension reduction, feature selection and confounding factor removal without explicit gene filtering. LCA has proved to be robust, accurate and powerful by comparison to multiple state-of-the-art computational methods on large-scale real and simulated scRNA-seq data. Importantly, LCA's ability to learn from representative subsets of the data provides scalability, thereby addressing a significant challenge for growing sample size in scRNA-seq data analysis. Citation Format: Changde Cheng, John Easton, Celeste Rosencrance, Yan Li, Bensheng Ju, Wenan Chen, Xiang Chen. LCA: A robust and scalable algorithm to reveal subtle diversity in large-scale single-cell RNA-Seq data [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 5297.