Skip to content

lixt314/scRNA-seq_notes

BranchesTags
 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

247 Commits
tools
tools
 
 
LICENSE
LICENSE
 
 
Pseudotime_tools.png
Pseudotime_tools.png
 
 
README.md
README.md
 
 
R_packages_for_scRNA-seq.pdf
R_packages_for_scRNA-seq.pdf
 
 

Repository files navigation

A continually expanding collection of scRNA-seq tools

Tools in each section are being resorted newest on top (previously, alphabetically). Check other bioinformatics resources and collections of links to various resources. Issues with suggestions and pull requests are welcome!

Table of content

Preprocessing pipelines

  • RAPIDS & Scanpy Single-Cell RNA-seq Workflow - real-time analysis of scRNA-seq data on GPU. Tweet

  • sceasy - An R package to convert different single-cell data formats to each other, supports Seurat, SingleCellExperiment, AnnData, Loom

  • kallistobus - fast pipeline for scRNA-seq processing. New BUS (Barcode, UMI, Set) format for storing and manipulating pseudoalignment results. Includes RNA velocity analysis. Python-based

  • PyMINEr - Python-based scRNA-seq processing pipeline. Cell type identification, detection of cell type-enriched genes, pathway analysis, co-expression networks and graph theory approaches to interpreting gene expression. Notes on methods: modified K++ clustering, automatic detection of the number of cell types, co-expression and PPI networks. Input: .txt or .hdf5 files. Detailed analysis of several pancreatic datasets

  • dropEst - pipeline for pre-processing, mapping, QCing, filtering, and quantifying droplet-based scRNA-seq datasets. Input - FASTQ or BAM, output - an R-readable molecular count matrix. Written in C++

  • zUMIs - scRNA-seq processing pipeline that handles barcodes and summarizes  UMIs using exonic or exonic + intronic mapped reads (improves clustering, DE detection). Adaptive downsampling of oversequenced libraries. STAR aligner, Rsubread::featureCounts counting UMIs in exons and introns.

  • bigSCale - scalable analytical framework to analyze large scRNA-seq datasets, UMIs or counts. Pre-clustering, convolution into iCells, final clustering, differential expression, biomarkers.Correlation metric for scRNA-seq data based on converting expression to Z-scores of differential expression. Robust to dropouts. Matlab implementation. Data, 1847 human neuronal progenitor cells

  • MAESTRO - Model-based AnalysEs of Single-cell Transcriptome and RegulOme - a comprehensive single-cell RNA-seq and ATAC-seq analysis suit built using snakemake

  • Scanpy - Python-based pipeline for preprocessing, visualization, clustering, pseudotime and trajectory inference, differential expression and network simulation

  • SEQC - Single-Cell Sequencing Quality Control and Processing Software, a general purpose method to build a count matrix from single cell sequencing reads, able to process data from inDrop, drop-seq, 10X, and Mars-Seq2 technologies

  • demuxlet - Introduces the ‘demuxlet’ algorithm, which enables genetic demultiplexing, doublet detection, and super-loading for droplet-based scRNA-seq. Recommended approach when samples have distinct genotypes

  • CALISTA - clustering, lineage reconstruction, transition gene identification, and cell pseudotime single cell transcriptional analysis. Analyses can be all or separate. Uses a likelihood-based approach based on probabilistic models of stochastic gene transcriptional bursts and random technical dropout events, so all analyses are compatible with each other. Input - a matrix of normalized, batch-removed log(RPKM) or log(TPM) or scaled UMIs. Methods detail statistical methodology. Matlab and R version

  • scPipe - A preprocessing pipeline for single cell RNA-seq data that starts from the fastq files and produces a gene count matrix with associated quality control information. It can process fastq data generated by CEL-seq, MARS-seq, Drop-seq, Chromium 10x and SMART-seq protocols. Modular, can swap tools like use different aligners

  • STAR alignment parameters: –outFilterType BySJout, –outFilterMultimapNmax 100, –limitOutSJcollapsed 2000000 –alignSJDBoverhangMin 8, –outFilterMismatchNoverLmax 0.04, –alignIntronMin 20, –alignIntronMax 1000000, –readFilesIn fastqrecords, –outSAMprimaryFlag AllBestScore, –outSAMtype BAM Unsorted. From Azizi et al., “Single-Cell Map of Diverse Immune Phenotypes in the Breast Tumor Microenvironment.”

Visualization pipelines

scATAC-seq

  • Notes by Ming Tang on scATAC-seq analysis, https://github.com/crazyhottommy/scATACseq-analysis-notes

  • Benchmarking of 10 scATAC-seq analysis methods (brief description of each in Methods) on 10 synthetic (various depth and noise levels) and 3 real datasets. scATAC technology overview, problems. Three clustering methods (K-means, Louvain, hierarchical clustering), adjusted Rand index, adjusted mutual information, homogeneity for benchmarking against gold-standard clustering, Residual Average Gini Index for benchmarking against gene markers (silver standard). SnapATAC, Cusanovich2018, cisTopic perform best overall. R code, Jupyter notebooks https://github.com/pinellolab/scATAC-benchmarking/

    • Chen, Huidong, Caleb Lareau, Tommaso Andreani, Michael E. Vinyard, Sara P. Garcia, Kendell Clement, Miguel A. Andrade-Navarro, Jason D. Buenrostro, and Luca Pinello. “Assessment of Computational Methods for the Analysis of Single-Cell ATAC-Seq Data.” Genome Biology 20, no. 1 (December 2019): 241. https://doi.org/10.1186/s13059-019-1854-5.

  • ArchR - R package for processing and analyzing single-cell ATAC-seq data. Compared to Signac and SnapATAC, has more functionality, faster. Input - BAM files. Efficient h5-based storage allows for large dataset processing. Doublet detection, genome-wide 500bp binning and peak identification, assignment to genes using best performing model, dimensionality reduction, clustering. Code to reproduce the paper, GitHub repo

  • SnapATAC - scATAC-seq pipeline for processing, clustering, and motif identification. Genome is binned into equal-size (5kb) windows, binarized with 1/0 for ATAC reads present/absent, Jaccard similarity between cells, normalized to account for sequencing depth (observed over expected method, two others), PCA on the matrix KNN graph and Louvain clustering to detect communities, tSNE or UMAP for visualization. Motif analysis and GREAT functional enrichment for each cluster. Outperforms ChromVAR, LSA, Cicero, Cis-Topic. Very fast, can be applied to ChIP-seq, scHi-C. https://github.com/r3fang/SnapATAC

    • Fang, Rongxin, Sebastian Preissl, Xiaomeng Hou, Jacinta Lucero, Xinxin Wang, Amir Motamedi, Andrew K. Shiau, et al. “Fast and Accurate Clustering of Single Cell Epigenomes Reveals Cis -Regulatory Elements in Rare Cell Types.” Preprint. Bioinformatics, April 22, 2019. https://doi.org/10.1101/615179.

  • Cicero - connect distal regulatory elements with target genes (covariance-based, graphical Lasso to compute a regularized covariance matrix) along pseudotime-ordered (Monocle2 or 3) scATAC-seq data. Optionally, adjusts for batch covariates. Applied to the analysis of skeletal myoblast differentiation, sciATAC-seq. R package, https://cole-trapnell-lab.github.io/cicero-release/

    • Pliner, Hannah A., Jonathan S. Packer, José L. McFaline-Figueroa, Darren A. Cusanovich, Riza M. Daza, Delasa Aghamirzaie, Sanjay Srivatsan, et al. “Cicero Predicts Cis-Regulatory DNA Interactions from Single-Cell Chromatin Accessibility Data.” Molecular Cell 71, no. 5 (September 2018): 858-871.e8. https://doi.org/10.1016/j.molcel.2018.06.044.

  • ChromVAR - scATAC-seq analysis. Identifying peaks, get a matrix of counts across aggregated peaks, tSNE for clustering, identifying motifs. Integrated with Seurat. https://github.com/GreenleafLab/chromVAR, https://greenleaflab.github.io/motifmatchr/index.html

    • Schep, Alicia N, Beijing Wu, Jason D Buenrostro, and William J Greenleaf. “ChromVAR: Inferring Transcription-Factor-Associated Accessibility from Single-Cell Epigenomic Data.” Nature Methods 14, no. 10 (August 21, 2017): 975–78. https://doi.org/10.1038/nmeth.4401.

  • scATAC-pro - pipeline for scATAC-seq mapping, QC, peak detection, clustering, TF and GO enrichment analysis, visualization (via VisCello). Compared with Scasat, Cellranger-atac. https://github.com/tanlabcode/scATAC-pro

    • “ScATAC-pro: A Comprehensive Workbench for Single-Cell Chromatin Accessibility Sequencing Data,” n.d., 24.

Quality control

Normalization

  • sctransform - using the Pearson residuals from regularized negative binomial regression where sequencing depth is utilized as a covariate to remove technical artifacts. Interfaces with Seurat.  https://cran.r-project.org/web/packages/sctransform/index.html

    • Hafemeister, Christoph, and Rahul Satija. “Normalization and Variance Stabilization of Single-Cell RNA-Seq Data Using Regularized Negative Binomial Regression.” BioRxiv, March 14, 2019. https://doi.org/10.1101/576827.

  • SCnorm - normalization for single-cell data. Quantile regression to estimate the dependence of transcript expression on sequencing depth for every gene. Genes with similar dependence are then grouped, and a second quantile regression is used to estimate scale factors within each group. Within-group adjustment for sequencing depth is then performed using the estimated scale factors to provide normalized estimates of expression. Good statistical methods description. https://www.biostat.wisc.edu/~kendzior/SCNORM/

    • Bacher, Rhonda, Li-Fang Chu, Ning Leng, Audrey P Gasch, James A Thomson, Ron M Stewart, Michael Newton, and Christina Kendziorski. “SCnorm: Robust Normalization of Single-Cell RNA-Seq Data.” Nature Methods 14, no. 6 (April 17, 2017): 584–86. https://doi.org/10.1038/nmeth.4263.

Batch effect, merging

  • Evaluation of 10 single-cell data integration methods and 4 preprocessing combinations on 77 batches of gene expression, chromatin accessibility, and simulated data (Table 1) in 9 integration tasks using 14 evaluation metrics. BBKNN, Scanorama, scVI perform well on complex tasks, Seurat performs well on simpler tasks but may eliminate biological signal. the use of Seurat v3 and Harmony is appropriate for simple integration tasks with distinct batch and biological structure. Batch in ATAC-seq is the most difficult to remove. Jupyter notebooks for full reproducibility https://github.com/theislab/scib

  • batchelor - Single-Cell Batch Correction Methods, by Aaron Lun. https://bioconductor.org/packages/devel/bioc/html/batchelor.html

  • iCellR - batch correction in scRNA-seq data. Combined Coverage Correction Alignment (CCCA) and Combined Principal Component Alignment (CPCA). CCCA - PCA into 30 dimensions, for each cell, take k=10 nearest neighbors, average gene expression, thus imputing the adjusted matrix. CPCA skips imputation, instead PCs themselves get averaged. Similar performance. Tested on nine PBMC datasets provided by the Broad institute to test batch effect. Outperforms MAGIC. https://cran.r-project.org/web/packages/iCellR/index.html, data in text and .rda formats at https://genome.med.nyu.edu/results/external/iCellR/data/

  • BERMUDA - Batch Effect ReMoval Using Deep Autoencoders, for scRNA-seq data. Requires batches to share at least one common cell type. Five step framework: 1) preprocessing, 2) clustering of cells in each batch individually, 3) identifying similar cell clusters across different batches, 4) removing batch effect by training an autoencoder, 5) further analysis of batch-corrected data. Tested on simulated (splatter) and experimental (10X genomics) data.https://github.com/txWang/BERMUDA

    • Wang, Tongxin, Travis S. Johnson, Wei Shao, Zixiao Lu, Bryan R. Helm, Jie Zhang, and Kun Huang. “BERMUDA: A Novel Deep Transfer Learning Method for Single-Cell RNA Sequencing Batch Correction Reveals Hidden High-Resolution Cellular Subtypes.” Genome Biology 20, no. 1 (December 2019). https://doi.org/10.1186/s13059-019-1764-6.

  • conos - joint analysis of scRNA-seq datasets through inter-sample mapping (mutual nearest-neighbor mapping) and constructing a joint graph. Analysis scripts: http://pklab.med.harvard.edu/peterk/conos/, R package:  https://github.com/hms-dbmi/conos

    • Barkas, Nikolas, Viktor Petukhov, Daria Nikolaeva, Yaroslav Lozinsky, Samuel Demharter, Konstantin Khodosevich, and Peter V. Kharchenko. “Joint Analysis of Heterogeneous Single-Cell RNA-Seq Dataset Collections.” Nature Methods, July 15, 2019. https://doi.org/10.1038/s41592-019-0466-z.

  • LIGER - R package for integrating and analyzing multiple single-cell datasets, across conditions, technologies (scRNA-seq and methylation), or species (human and mouse). Integrative nonnegative matrix factorization (W and H matrices), dataset-specific and shared patterns (metagenes, matrix H). Graphs of factor loadings onto these patterns (shared factor neighborhood graph), then comparing patterns. Alignment and agreement metrics to assess performance, LIGER outperforms Seurat on agreement. Analysis of published blood cells, brain. Human/mouse brain data at https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE126836.   https://github.com/MacoskoLab/liger

    • Welch, Joshua D., Velina Kozareva, Ashley Ferreira, Charles Vanderburg, Carly Martin, and Evan Z. Macosko. “Single-Cell Multi-Omic Integration Compares and Contrasts Features of Brain Cell Identity.” Cell 177, no. 7 (June 13, 2019): 1873-1887.e17. https://doi.org/10.1016/j.cell.2019.05.006.

  • scMerge - R package for batch effect removal and normalizing of multipe scRNA-seq datasets. fastRUVIII batch removal method. Tested on 14 datasets, compared with scran, MNN, ComBat, Seurat, ZINB-WaVE using Silhouette, ARI - better separation of clusters, pseudotime reconstruction. https://github.com/SydneyBioX/scMerge/

    • Lin, Yingxin, Shila Ghazanfar, Kevin Wang, Johann A. Gagnon-Bartsch, Kitty K. Lo, Xianbin Su, Ze-Guang Han, et al. “ScMerge: Integration of Multiple Single-Cell Transcriptomics Datasets Leveraging Stable Expression and Pseudo-Replication,” September 12, 2018. https://doi.org/10.1101/393280.

  • MNN - mutual nearest neighbors method for single-cell batch correction. Assumptions: MNN exist between batches, batch is orthogonal to the biology. Cosine normalization, Euclidean distance, a pair-specific barch-correction vector as a vector difference between the expression profiles of the paired cells using selected genes of interest and hypervariable genes. Supplementary note 5 - algorithm. mnnCorrect function in the scran package https://bioconductor.org/packages/release/bioc/html/scran.html. Code for paper https://github.com/MarioniLab/MNN2017/

    • Haghverdi, Laleh, Aaron T L Lun, Michael D Morgan, and John C Marioni. “Batch Effects in Single-Cell RNA-Sequencing Data Are Corrected by Matching Mutual Nearest Neighbors.” Nature Biotechnology, April 2, 2018. https://doi.org/10.1038/nbt.4091.

  • scLVM - a modelling framework for single-cell RNA-seq data that can be used to dissect the observed heterogeneity into different sources and remove the variation explained by latent variables. Can correct for the cell cycle effect. Applied to naive T cells differentiating into TH2 cells. https://github.com/PMBio/scLVM

    • Buettner, Florian, Kedar N Natarajan, F Paolo Casale, Valentina Proserpio, Antonio Scialdone, Fabian J Theis, Sarah A Teichmann, John C Marioni, and Oliver Stegle. “Computational Analysis of Cell-to-Cell Heterogeneity in Single-Cell RNA-Sequencing Data Reveals Hidden Subpopulations of Cells.” Nature Biotechnology 33, no. 2 (March 2015): 155–60. https://doi.org/10.1038/nbt.3102.
    • Buettner, Florian, Naruemon Pratanwanich, Davis J. McCarthy, John C. Marioni, and Oliver Stegle. “F-ScLVM: Scalable and Versatile Factor Analysis for Single-Cell RNA-Seq.” Genome Biology 18, no. 1 (December 2017). https://doi.org/10.1186/s13059-017-1334-8. - f-scLVM - factorial single-cell latent variable model guided by pathway annotations to infer interpretable factors behind heterogeneity. PCA components are annotated by correlated genes and their enrichment in pathways. Docomposition of the original gene expression matrix to a sum of annotated, unannotated, and confounding components. Applied to their own naive T to TH2 cells, mESCs, reanalyzed 3005 neuronal cells. Simulated data. https://github.com/bioFAM/slalom

Imputation

  • Assessment of 18 scRNA-seq imputation methods on similarity with the original data and downstream analyses - differential expression, unsupervised clustering, pseudotime trajectory inference. The majority of methods do not improve clustering, trajectory inference. Overall performance highly variable, depends on many factors, data type. MAGIC, kNN-smoothing, SAVER performed on average better. Code to get the data and perform all analyses https://github.com/Winnie09/imputationBenchmark

    • Hou, Wenpin, Zhicheng Ji, Hongkai Ji, and Stephanie C. Hicks. “A Systematic Evaluation of Single-Cell RNA-Sequencing Imputation Methods.” Preprint. Genomics, January 30, 2020. https://doi.org/10.1101/2020.01.29.925974.

  • netNMF-sc - scRNA-seq nonnegative matrix factorization for imputation and dimensionality reduction for improved clustering. Uses gene-gene interaction network to constrain W gene matrix on prior knowledge (graph regularized NMF). Added penalization for dropouts. Tested on simulated and experimental data, compared with several imputation and clustering methods. https://github.com/raphael-group/netNMF-sc

    • Elyanow, Rebecca, Bianca Dumitrascu, Barbara E Engelhardt, and Benjamin J Raphael. “NetNMF-Sc: Leveraging Gene-Gene Interactions for Imputation and Dimensionality Reduction in Single-Cell Expression Analysis.” BioRxiv, February 8, 2019. https://doi.org/10.1101/544346.

  • SCRABBLE - scRNA-seq imputation constraining on bulk RNA-seq data. Matrix regularzation optimizing a three-term objective function. Compared with DrImpute, scImpute, MAGIC, VIPER on simulated and real data. Datasets: https://github.com/tanlabcode/SCRABBLE_PAPER. R and Matlab implementation. https://github.com/tanlabcode/SCRABBLE

    • Peng, Tao, Qin Zhu, Penghang Yin, and Kai Tan. “SCRABBLE: Single-Cell RNA-Seq Imputation Constrained by Bulk RNA-Seq Data.” Genome Biology 20, no. 1 (December 2019): 88. https://doi.org/10.1186/s13059-019-1681-8.

  • Deepimpute - scRNA-seq imputation using deep neural networks. Sub-networks, each processes up to 512 genes needed to be imputed. Four layers: Input - dense (ReLU activation) - 20% dropout - output. MSE as loss function. Outperforms MAGIC, DrImpute, ScImpute, SAVER, VIPER, and DCA on multiple metrics (PCC, several clustering metrics). Using 9 datasets. https://github.com/lanagarmire/DeepImpute

    • Arisdakessian, Cédric, Olivier Poirion, Breck Yunits, Xun Zhu, and Lana X. Garmire. “DeepImpute: An Accurate, Fast, and Scalable Deep Neural Network Method to Impute Single-Cell RNA-Seq Data.” Genome Biology 20, no. 1 (December 2019): 211. https://doi.org/10.1186/s13059-019-1837-6.

  • scHinter - imputation for small-size scRNA-seq datasets. Three modules: voting-based ensemble distance for learning cell-cell similarity, a SMOTE-based random interpolation module for imputing dropout events, and a hierarchical model for multi-layer random interpolation.  https://github.com/BMILAB/scHinter, RNA-seq blog

    • Ye, Pengchao, Wenbin Ye, Congting Ye, Shuchao Li, Lishan Ye, Guoli Ji, and Xiaohui Wu. “ScHinter: Imputing Dropout Events for Single-Cell RNA-Seq Data with Limited Sample Size.” Edited by Inanc Birol. Bioinformatics, August 8, 2019. https://doi.org/10.1093/bioinformatics/btz627.

  • ENHANCE, an algorithm that denoises single-cell RNA-Seq data by first performing nearest-neighbor aggregation and then inferring expression levels from principal components. Variance-stabilizing normalization of the data before PCA. Implements its own simulation procedure for simulating sampling noise. Outperforms MAGIC, SAVER, ALRA. Python: https://github.com/yanailab/enhance, and R implementation: https://github.com/yanailab/enhance-R

    • Wagner, Florian, Dalia Barkley, and Itai Yanai. “ENHANCE: Accurate Denoising of Single-Cell RNA-Seq Data.” Preprint. Bioinformatics, June 3, 2019. https://doi.org/10.1101/655365.

  • scRMD - dropout imputation in scRNA-seq via robust matrix decomposition into true expression matrix (further decomposed into a matrix of means and gene's random deviation from its mean) minus dropout matrix plus error matrix. A function to estimate the matrix of means and dropouts. Comparison with MAGIC, scImpute. https://github.com/ChongC1990/scRMD

    • Chen, Chong, Changjing Wu, Linjie Wu, Yishu Wang, Minghua Deng, and Ruibin Xi. “ScRMD: Imputation for Single Cell RNA-Seq Data via Robust Matrix Decomposition,” November 4, 2018. https://doi.org/10.1101/459404.

  • netSmooth - network diffusion-based method that uses priors for the covariance structure of gene expression profiles to smooth scRNA-seq experiments. Incorporates prior knowledge (i.e. protein-protein interaction networks) for imputation. Note that dropout applies to whole transcriptome. Compared with MAGIC, scImpute. Improves clustering, biological interpretation. https://github.com/BIMSBbioinfo/netSmooth

  • DCA - A deep count autoencoder network to denoise scRNA-seq data. Zero-inflated negative binomial model. Current approaches - scimpute, MAGIC, SAVER. Benchmarking by increased correlation between bulk and scRNA-seq data, between protein and RNA levels, between key regulatory genes, better DE concordance in bulk and scRNA-seq, improved clustering, https://github.com/theislab/dca

    • Eraslan, Gökcen, Lukas M. Simon, Maria Mircea, Nikola S. Mueller, and Fabian J. Theis. “Single Cell RNA-Seq Denoising Using a Deep Count Autoencoder,” April 13, 2018. https://doi.org/10.1101/300681.

  • kNN-smoothing of scRNA-seq data, aggregates information from similar cells, improves signal-to-noise ratio. Based on observation that gene expression in technical replicates are Poisson distributed. Freeman-Tukey transform to minimize variability of low expressed genes. Tested using real and simulated data. Improves clustering, PCA, Selection of k is critical, discussed.https://github.com/yanailab/knn-smoothing

    • Wagner, Florian, Yun Yan, and Itai Yanai. “K-Nearest Neighbor Smoothing for High-Throughput Single-Cell RNA-Seq Data.” BioRxiv, April 9, 2018. https://doi.org/10.1101/217737.

  • scimpute - imputation of scRNA-seq data. Methodology: 1) Determine K subpopulations using PCA, remove outliers; 2) Mixture model of gene i in subpopulation k as gamma and normal distributions, estimate dropout probability d; 3) Impute dropout values by splitting the subpopulation into A (dropout larger than threshold t) and B (smaller). Information from B is used to impute A. Better than MAGIC, SAVER. https://github.com/Vivianstats/scImpute

    • Li, Wei Vivian, and Jingyi Jessica Li. “An Accurate and Robust Imputation Method ScImpute for Single-Cell RNA-Seq Data.” Nature Communications 9, no. 1 (08 2018): 997. https://doi.org/10.1038/s41467-018-03405-7.

  • LATE (Learning with AuToEncoder) to imputescRNA-seq data. TRANSLATE (TRANSfer learning with LATE) uses reference (sc)RNA-seq dataset to learn initial parameter estimates. TensorFlow implementation for GPU and CPU. ReLu as an activation function. Various optimization techniques. Comparison with MAGIC, scVI, DCA, SAVER. Links to data. https://github.com/audreyqyfu/LATE 

    • Badsha, Md. Bahadur, Rui Li, Boxiang Liu, Yang I. Li, Min Xian, Nicholas E. Banovich, and Audrey Qiuyan Fu. “Imputation of Single-Cell Gene Expression with an Autoencoder Neural Network.” BioRxiv, January 1, 2018, 504977. https://doi.org/10.1101/504977.

  • MAGIC - Markov Affinity-based Graph Imputation of Cells. Only ~5-15% of scRNA-seq data is non-zero, the rest are drop-outs. Use the diffusion operator to discover the manifold structure and impute gene expression. Detailed methods description. In real (bone marrow and retinal bipolar cells) and synthetic datasets, Imputed scRNA-seq data clustered better, enhances gene interactions, restores expression of known surface markers, trajectories. scRNA-seq data is preprocessed by library size normalization and PCA (to retain 70% of variability). Comparison with SVD-based low-rank data approximation (LDA) and Nuclear-Norm-based Matrix Completion (NNMC). https://github.com/KrishnaswamyLab/MAGIC, https://cran.r-project.org/web/packages/Rmagic/index.html

    • Dijk, David van, Juozas Nainys, Roshan Sharma, Pooja Kathail, Ambrose J Carr, Kevin R Moon, Linas Mazutis, Guy Wolf, Smita Krishnaswamy, and Dana Pe’er. “MAGIC: A Diffusion-Based Imputation Method Reveals Gene-Gene Interactions in Single-Cell RNA-Sequencing Data,” February 25, 2017. https://doi.org/10.1101/111591.

Dimensionality reduction

  • CIDR - Clustering through Imputation and Dimensionality Reduction. Impute dropouts. Explicitly deconvolve Euclidean distance into distance driven by complete, partially complete, and dropout pairs. Principal Coordinate Analysis. https://github.com/VCCRI/CIDR

    • Lin, Peijie, Michael Troup, and Joshua W. K. Ho. “CIDR: Ultrafast and Accurate Clustering through Imputation for Single-Cell RNA-Seq Data.” Genome Biology 18, no. 1 (December 2017). https://doi.org/10.1186/s13059-017-1188-0.

  • RobustAutoencoder - Autoencoder and robust PCA for gene expression representation, robust to outliers. Main idea - split the input data X into two parts, L (reconstructed data) and S (outliers and noise). Grouped "l2,1" norm - an l2 regularizer within a group and then an l1 regularizer between groups. Iterative procedure to obtain L and S. TensorFlow implementation. https://github.com/zc8340311/RobustAutoencoder

    • Zhou, Chong, and Randy C. Paffenroth. “Anomaly Detection with Robust Deep Autoencoders.” In Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD ’17, 665–74. Halifax, NS, Canada: ACM Press, 2017. https://doi.org/10.1145/3097983.3098052.

  • SAUCIE - deep neural network with regularization on layers to improve interpretability. Denoising, batch removal, imputation, visualization of low-dimensional representation. Extensive comparison on simulated and real data. https://github.com/KrishnaswamyLab/SAUCIE

    • Amodio, Matthew, David van Dijk, Krishnan Srinivasan, William S Chen, Hussein Mohsen, Kevin R Moon, Allison Campbell, et al. “Exploring Single-Cell Data with Deep Multitasking Neural Networks,” August 27, 2018. https://doi.org/10.1101/237065.

  • ZIFA - Zero-inflated dimensionality reduction algorithm for single-cell data. Single-cell dimensionality reduction. Model dropout rate as double exponential, give less weights to these counts. EM algorithm that incorporates imputation step for the expected gene expression level of drop-outs. https://github.com/epierson9/ZIFA

    • Pierson, Emma, and Christopher Yau. “ZIFA: Dimensionality Reduction for Zero-Inflated Single-Cell Gene Expression Analysis.” Genome Biology 16 (November 2, 2015): 241. https://doi.org/10.1186/s13059-015-0805-z.

  • ZINB-WAVE - Zero-inflated negative binomial model for normalization, batch removal, and dimensionality reduction. Extends the RUV model with more careful definition of "unwanted" variation as it may be biological. Good statistical derivations in Methods. Refs to real and simulated scRNA-seq datasets. https://bioconductor.org/packages/release/bioc/html/zinbwave.html

    • Risso, Davide, Fanny Perraudeau, Svetlana Gribkova, Sandrine Dudoit, and Jean-Philippe Vert. “ZINB-WaVE: A General and Flexible Method for Signal Extraction from Single-Cell RNA-Seq Data.” BioRxiv, January 1, 2017. https://doi.org/10.1101/125112.

  • UMAP (Uniform Manifold Approximation and Projection) - dimensionality reduction using machine learning. Detailed statistical framework. Compared with t-SNE, better preserves global structure. http://github.com/lmcinnes/umap. R implementation: https://github.com/jlmelville/uwot

    • McInnes, Leland, and John Healy. “UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction.” ArXiv:1802.03426 [Cs, Stat], February 9, 2018. http://arxiv.org/abs/1802.03426.

  • VASC - deep variational autoencoder for scRNA-seq data for dimensionality reduction and visualization. Tested on twenty datasets vs PCA, tSNE, ZIFA, and SIMLR. Four metrics to assess clustering performance: NMI (normalized mutual information score), ARI (adjusted rand index), HOM (homogeneity) and COM (completeness). No filtering, only log transformation. Keras implementation. Datasets https://hemberg-lab.github.io/scRNA.seq.datasets/, and the code https://github.com/wang-research/VASC

    • Wang, Dongfang, and Jin Gu. “VASC: Dimension Reduction and Visualization of Single Cell RNA Sequencing Data by Deep Variational Autoencoder,” October 6, 2017. https://doi.org/10.1101/199315.

Clustering

  • Recommendations to properly use t-SNE on large omics datasets (scRNA-seq in particular) to preserve global geometry. Overview of t-SNE, PCA, MDS, UMAP, their similarities, differences, strengths and weaknesses. PCA initialization (first two components are OK), a high learning rate of n/12, and multi-scale similarity kernels. For very large data, increase exagerration. Strategies to align new points on an existing t-SNE plot, aligning two t-SNE visualizations. Extremely fast implementation is FIt-SNE, https://github.com/KlugerLab/FIt-SNE. Code to illustrate the use of t-SNE: https://github.com/berenslab/rna-seq-tsne

  • Performance evaluation of 14 scRNA-seq clustering algorithms using nine experimental and three simulated datasets. SC3 and Seurat perform best overall. Normalized Shannon entropy, adjusted Rand index for performance evaluation. Ensemble clustering doesn't help. Scripts https://github.com/markrobinsonuzh/scRNAseq_clustering_comparison and a data package for clustering benchmarking with preprocessed and experimental scRNA-seq datasets https://bioconductor.org/packages/release/data/experiment/html/DuoClustering2018.html

    • Duò, Angelo, Mark D. Robinson, and Charlotte Soneson. “A Systematic Performance Evaluation of Clustering Methods for Single-Cell RNA-Seq Data.” F1000Research 7 (September 10, 2018): 1141. https://doi.org/10.12688/f1000research.15666.2.

  • BAMM-SC - scRNA-seq clustering. A Bayesian hierarchical Dirichlet multinomial mixture model, accounts for batch effect, operates on raw counts. Outperforms K-means, TSCAN, Seurat corrected for batch using MNN or CCA in simulated and experimental settings. https://github.com/CHPGenetics/BAMMSC

    • Sun, Zhe, Li Chen, Hongyi Xin, Yale Jiang, Qianhui Huang, Anthony R. Cillo, Tracy Tabib, et al. “A Bayesian Mixture Model for Clustering Droplet-Based Single-Cell Transcriptomic Data from Population Studies.” Nature Communications 10, no. 1 (December 2019): 1649. https://doi.org/10.1038/s41467-019-09639-3.

  • scClustViz - assessment of scRNA-seq clustering using differential expression (Wilcoxon test) as a guide. Testing for two differences: difference in detection rate (dDR) and log2 gene expression ratio (logGER). Two hypothesis testing: one cluster vs. all, each cluster vs. another cluster. accepts SincleCellExperiment and Seurat objects (log2-transformed data), needs a data frame with different cluster assignments. Analysis within R, save as RData, visualize results in R Shiny app. https://github.com/BaderLab/scClustViz

  • Spectrum - a spectral clustering method for single- or multi-omics datasets. Self-tuning kernel that adapts to local density of the graph. Tensor product graph data integration method. Implementation of fast spectral clustering method (single dataset only). Finds optimal number of clusters using eigenvector distribution analysis. References to previous methods. Excellent methods description. Compared with M3C, CLEST, PINSplus, SNF, iClusterPlus, CIMLR, MUDAN.  https://github.com/crj32/spectrum_manuscript, https://cran.r-project.org/web/packages/Spectrum/index.html

    • John, Christopher R., David Watson, Michael R. Barnes, Costantino Pitzalis, and Myles J. Lewis. “Spectrum: Fast Density-Aware Spectral Clustering for Single and Multi-Omic Data.” Bioinformatics (Oxford, England), September 10, 2019. https://doi.org/10.1093/bioinformatics/btz704.

  • PanoView - scRNA-seq iterative clustering in an evolving 3D PCA space, Ordering Local Maximum by Convex hull (OLMC) to identify clusters of varying density. PCA on most variable genes, finding most optimal largest cluster within first 3 PCs, repeat PCA for the remaining cells etc. Tested on multiple simulated and experimental scRNA-seq datasets, compared with 9 methods, the Adjusted Rand Index as performance metric. https://github.com/mhu10/scPanoView

    • Hu, Ming-Wen, Dong Won Kim, Sheng Liu, Donald J. Zack, Seth Blackshaw, and Jiang Qian. “PanoView: An Iterative Clustering Method for Single-Cell RNA Sequencing Data.” Edited by Qing Nie. PLOS Computational Biology 15, no. 8 (August 30, 2019): e1007040. https://doi.org/10.1371/journal.pcbi.1007040.

  • TooManyCells - divisive hierarchical spectral clustering of scRNA-seq data. Uses truncated singular vector decomposition to bipartition the cells. Newman-Girvain modularity Q to assess whether bipartition is significant or should be stopped. BirchBeer visualization. Outperforms Phenograph, Seurat, Cellranger, Monocle, the latter is second in performance. Excels for rare populations. Normalization marginally affects performance. https://github.com/GregorySchwartz/tooManyCellsR and  https://github.com/faryabiLab/birch-beer

    • Schwartz, Gregory W, Jelena Petrovic, Maria Fasolino, Yeqiao Zhou, Stanley Cai, Lanwei Xu, Warren S Pear, Golnaz Vahedi, and Robert B Faryabi. “TooManyCells Identifies and Visualizes Relationships of Single-Cell Clades.” BioRxiv, January 13, 2019. https://doi.org/10.1101/519660.

  • PHATE (Potential of Heat-diffusion for Affinity-based Transition Embedding) - low-dimensional embedding, denoising, and visualization, applicable to scRNA-seq, microbiome, SNP, Hi-C (as affinity matrices) and other data. Preserves biological structures and branching better than PCA, tSNE, diffusion maps. Robust to noise and subsampling. Detailed methods description and graphical representation of the algorithm. https://github.com/KrishnaswamyLab/PHATE, Tweetorial

    • Moon, Kevin R., David van Dijk, Zheng Wang, Scott Gigante, Daniel Burkhardt, William Chen, Antonia van den Elzen, et al. “Visualizing Transitions and Structure for Biological Data Exploration,” June 28, 2018. https://doi.org/10.1101/120378.

  • FIt-SNE - accelerated version of t-SNE clustering for visualizing thousands/milions of cells. https://github.com/KlugerLab/FIt-SNE. t-SNE-Heatmaps - discretized t-SNE clustering representation as a heatmap. https://github.com/KlugerLab/t-SNE-Heatmaps. Detailed methods, and further stats at https://gauss.math.yale.edu/~gcl22/blog/numerics/low-rank/t-sne/2018/01/11/low-rank-kernels.html

    • Linderman, George C., Manas Rachh, Jeremy G. Hoskins, Stefan Steinerberger, and Yuval Kluger. “Fast Interpolation-Based t-SNE for Improved Visualization of Single-Cell RNA-Seq Data.” Nature Methods, February 11, 2019. https://doi.org/10.1038/s41592-018-0308-4.

  • MetaCell - partitioning scRNA-seq data into metacells - disjoint and homogeneous/compact groups of cells exhibiting only sampling variance. Most variable genes to cell-to-cell similarity matrix (PCC on  to Knn similarity graph that is partitioned by bootstrapping to obtain subgraphs. Tested on several 10X datasets, https://support.10xgenomics.com/single-cell-gene-expression/datasets. https://bitbucket.org/tanaylab/metacell/src/default/

    • Baran, Yael, Arnau Sebe-Pedros, Yaniv Lubling, Amir Giladi, Elad Chomsky, Zohar Meir, Michael Hoichman, Aviezer Lifshitz, and Amos Tanay. “MetaCell: Analysis of Single Cell RNA-Seq Data Using k-NN Graph Partitions.” BioRxiv, January 1, 2018, 437665. https://doi.org/10.1101/437665.

  • clusterExperiment R package for scRNA-seq data visualization. Resampling-based Sequential Ensemble Clustering (RSEC) method. clusterMany - makeConsensus - makeDendrogram - mergeClusters pipeline. Biomarker detection by differential expression analysis between clusters. Visualization as heatmaps, https://bioconductor.org/packages/release/bioc/html/clusterExperiment.html

    • Risso, Davide, Liam Purvis, Russell B. Fletcher, Diya Das, John Ngai, Sandrine Dudoit, and Elizabeth Purdom. “ClusterExperiment and RSEC: A Bioconductor Package and Framework for Clustering of Single-Cell and Other Large Gene Expression Datasets.” Edited by Aaron E. Darling. PLOS Computational Biology 14, no. 9 (September 4, 2018): e1006378. https://doi.org/10.1371/journal.pcbi.1006378.

  • Bisquit - a Bayesian clustering and normalization method. https://github.com/sandhya212/BISCUIT_SingleCell_IMM_ICML_2016

    • Azizi, Elham, Ambrose J. Carr, George Plitas, Andrew E. Cornish, Catherine Konopacki, Sandhya Prabhakaran, Juozas Nainys, et al. “Single-Cell Map of Diverse Immune Phenotypes in the Breast Tumor Microenvironment.” Cell, June 2018. https://doi.org/10.1016/j.cell.2018.05.060.

  • scVAE - Variational auroencoder frameworks for modelling raw RNA-seq counts, denoising the data to improve biologically plausible grouping in scRNA-seq data. Improvement in Rand index. https://github.com/chgroenbech/scVAE

    • Grønbech, Christopher Heje, Maximillian Fornitz Vording, Pascal N Timshel, Casper Kaae Sønderby, Tune Hannes Pers, and Ole Winther. “ScVAE: Variational Auto-Encoders for Single-Cell Gene Expression Data,” May 16, 2018. https://doi.org/10.1101/318295.

  • Conos - clustering of scRNA-seq samples by joint graph construction. Seurat or pagoda2 for data preprocessing, selection of hypervariable genes, initial clustering (KNN, or dimensionality reduction), then joint clustering. R package, https://github.com/hms-dbmi/conos

    • Barkas, Nikolas, Viktor Petukhov, Daria Nikolaeva, Yaroslav Lozinsky, Samuel Demharter, Konstantin Khodosevich, and Peter V Kharchenko. “Wiring Together Large Single-Cell RNA-Seq Sample Collections.” BioRxiv, January 1, 2018. https://doi.org/10.1101/460246.

  • SIMLR - scRNA-seq dimensionality reduction, clustering, and visualization based on multiple kernel-learned distance metric. Comparison with PCA, t-SNE, ZIFA. Seven datasets. R and Matlab implementation. https://github.com/BatzoglouLabSU/SIMLR

    • Wang, Bo, Junjie Zhu, Emma Pierson, Daniele Ramazzotti, and Serafim Batzoglou. “Visualization and Analysis of Single-Cell RNA-Seq Data by Kernel-Based Similarity Learning.” Nature Methods 14, no. 4 (April 2017): 414–16. https://doi.org/10.1038/nmeth.4207.

  • SC3 - single-cell clustering. Multiple clustering iterations, consensus matrix, then hierarhical clustering. Benchmarking against other methods. https://bioconductor.org/packages/release/bioc/html/SC3.html

    • Kiselev, Vladimir Yu, Kristina Kirschner, Michael T Schaub, Tallulah Andrews, Andrew Yiu, Tamir Chandra, Kedar N Natarajan, et al. “SC3: Consensus Clustering of Single-Cell RNA-Seq Data.” Nature Methods 14, no. 5 (March 27, 2017): 483–86. https://doi.org/10.1038/nmeth.4236.

  • destiny - R package for diffusion maps-based visualization of single-cell data. https://bioconductor.org/packages/release/bioc/html/destiny.html

    • Haghverdi, Laleh, Florian Buettner, and Fabian J. Theis. “Diffusion Maps for High-Dimensional Single-Cell Analysis of Differentiation Data.” Bioinformatics 31, no. 18 (September 15, 2015): 2989–98. https://doi.org/10.1093/bioinformatics/btv325. - Introduction of other methods, Table 1 compares them. Methods details. Performance is similar to PCA and tSNE.

  • PhenoGraph - discovers subpopulations in scRNA-seq data. High-dimensional space is modeled as a nearest-neighbor graph, then the Louvain community detection algorithm. No assumptions about the size, number, or form of subpopulations. https://github.com/jacoblevine/PhenoGraph

    • Levine, Jacob H., Erin F. Simonds, Sean C. Bendall, Kara L. Davis, El-ad D. Amir, Michelle D. Tadmor, Oren Litvin, et al. “Data-Driven Phenotypic Dissection of AML Reveals Progenitor-like Cells That Correlate with Prognosis.” Cell 162, no. 1 (July 2015): 184–97. https://doi.org/10.1016/j.cell.2015.05.047.

  • SNN-Cliq - shared nearest neighbor clustering of scRNA-seq data, represented as a graph. Similarity between two data points based on the ranking of their shared neighborhood. Automatically determine the number of clusters, accomodates different densities and shapes. Compared with K-means and DBSCAN using Purity, Adjusted Rand Indes, F1-score. Matlab, Python, R implementation. http://bioinfo.uncc.edu/SNNCliq/

    • Xu, Chen, and Zhengchang Su. “Identification of Cell Types from Single-Cell Transcriptomes Using a Novel Clustering Method.” Bioinformatics (Oxford, England) 31, no. 12 (June 15, 2015): 1974–80. https://doi.org/10.1093/bioinformatics/btv088.

  • viSNE - the Barnes-Hut implementation of the t-SNE algorithm, improved and tailored for the analysis of single-cell data. Details of tSNE https://www.denovosoftware.com/site/manual/visne.htm, and the Rtsne R package https://cran.r-project.org/web/packages/Rtsne/index.html

    • Amir, El-ad David, Kara L Davis, Michelle D Tadmor, Erin F Simonds, Jacob H Levine, Sean C Bendall, Daniel K Shenfeld, Smita Krishnaswamy, Garry P Nolan, and Dana Pe’er. “ViSNE Enables Visualization of High Dimensional Single-Cell Data and Reveals Phenotypic Heterogeneity of Leukemia.” Nature Biotechnology 31, no. 6 (June 2013): 545–52. https://doi.org/10.1038/nbt.2594.

  • celda - CEllular Latent Dirichlet Allocation. Simultaneous clustering of cells into subpopulations and genes into transcriptional states. https://github.com/compbiomed/celda. Tutorials will be at https://github.com/compbiomed/celda_tutorials. No preprint yet.

Spatial inference

Time, trajectory inference

  • A collection of 57 trajectory inference methods, https://github.com/dynverse/dynmethods#list-of-included-methods

    • Saelens, Wouter, Robrecht Cannoodt, Helena Todorov, and Yvan Saeys. “A Comparison of Single-Cell Trajectory Inference Methods: Towards More Accurate and Robust Tools,” March 5, 2018. https://doi.org/10.1101/276907. - Review of trajectory 29 inference methods for single-cell RNA-seq (out of 57 methods collected). Slingshot, TSCAN and Monocle DDRTree perform best overall. https://github.com/dynverse/dynverse

  • Single-cell RNA-seq pseudotime estimation algorithms, by Anthony Gitter. References and descriptions of many algorithms. https://github.com/agitter/single-cell-pseudotime

  • Table of comparison of 11 trajectory inference methods, Supplementary Table 1 from Huidong Chen et al., “Single-Cell Trajectories Reconstruction, Exploration and Mapping of Omics Data with STREAM,” Nature Communications 10, no. 1 (April 23, 2019): 1903, https://doi.org/10.1038/s41467-019-09670-4.

  • LACE - R package for Longitudinal Analysis of Cancer Evolution using mutations called from scRNA-seq data. Input - binary matrix with 1 indicating somatic variants, called using GATK. Boolean matrix factorization solved via exhaustive search or via MCMC. Output - the maximum likelihood clonal tree describing the longitudinal evolution of a tumor. Compared with CALDER, SCITE, TRaIT. https://github.com/BIMIB-DISCo/LACE

    • Ramazzotti, Daniele, Fabrizio Angaroni, Davide Maspero, Gianluca Ascolani, Isabella Castiglioni, Rocco Piazza, Marco Antoniotti, and Alex Graudenzi. “Longitudinal Cancer Evolution from Single Cells.” Preprint. Bioinformatics, January 15, 2020. https://doi.org/10.1101/2020.01.14.906453.

  • Tempora - pseudotime reconstruction for scRNA-seq data, considering time series information. Matrix of scRNA-seq gene expression to clusters to pathways (single-cell enrichment), to graph (ARACNE), incorporate time scores, refine cell types. Outperforms Monocle2, TSCAN. https://github.com/BaderLab/Tempora

    • Tran, Thinh N., and Gary D. Bader. “Tempora: Cell Trajectory Inference Using Time-Series Single-Cell RNA Sequencing Data.” Preprint. Bioinformatics, November 18, 2019. https://doi.org/10.1101/846907.

  • STREAM (Single-cell Trajetrories Reconstruction, Exploration And Mapping) - reconstruction of pseudotime trajectory with branching from scRNA-seq or scATAC-seq data. Principal graph, modified locally linear embedding, Elastic Principal Graph (previously published). Illustrated on several datasets. Compared with ten methods (Monocle2, scTDA, Wishbone, TSCAN, SLICER, DPT, GPFates, Mpath, SCUBA, PHATE). Web-version, visualization of published datasets: http://stream.pinellolab.org/, Jupyter notebooks, Docker container: https://github.com/pinellolab/STREAM

    • Chen, Huidong, Luca Albergante, Jonathan Y. Hsu, Caleb A. Lareau, Giosuè Lo Bosco, Jihong Guan, Shuigeng Zhou, et al. “Single-Cell Trajectories Reconstruction, Exploration and Mapping of Omics Data with STREAM.” Nature Communications 10, no. 1 (April 23, 2019): 1903. https://doi.org/10.1038/s41467-019-09670-4.

  • Monocle 2 - Reversed graph embedding (DDRTree), finding low-dimensional mapping of differential genes while learning the graph in this reduced space. Allows for the selection of root. Compared with Monocle 1, Wishbone, Diffusion Pseudotime, SLICER. Code, https://github.com/cole-trapnell-lab/monocle-release, analysis, https://github.com/cole-trapnell-lab/monocle2-rge-paper

    • Qiu, Xiaojie, Qi Mao, Ying Tang, Li Wang, Raghav Chawla, Hannah A Pliner, and Cole Trapnell. “Reversed Graph Embedding Resolves Complex Single-Cell Trajectories.” Nature Methods 14, no. 10 (August 21, 2017): 979–82. https://doi.org/10.1038/nmeth.4402.

-Slingshot - Inferring multiple developmental lineages from single-cell gene expression. Clustering by gene expression, then inferring cell lineage as an ordered set of clusters -minimum spanning tree through the clusters using Mahalanobis distance. Initial state and terminal state specification. Principal curves to draw a path through the gene expression space of each lineage. https://github.com/kstreet13/slingshot - Street, Kelly, Davide Risso, Russell B Fletcher, Diya Das, John Ngai, Nir Yosef, Elizabeth Purdom, and Sandrine Dudoit. “Slingshot: Cell Lineage and Pseudotime Inference for Single-Cell Transcriptomics.” BioRxiv, January 1, 2017. https://doi.org/10.1186/s12864-018-4772-0.

  • SCITE - a stochastic search algorithm to identify the evolutionary history of a tumor from mutation patterns in scRNA-seq data. MCMC to compute the maximum-likelihood mutation history. Accounts for noise and dropouts. Input - Boolean mutation matrix, output - maximum-likelihood-inferred mutation tree. Compared with Kim & Simon approach, BitPhylogeny. https://github.com/cbg-ethz/SCITE

  • DPT - diffusion pseudotime, arrange cells in the pseudotemporal order. Random-walk-based distance that is computed based on Euclidean distance in the diffusion map space. Weighted nearest neighborhood of the data, probabilities of transitioning to each other cell using random walk, DTP is the euclidean distance between the two vectors, stored in a transition matrix. Robust to noise and sparsity. Method compared with Monocle, Wishbone, Wanderlust. https://theislab.github.io/destiny/index.html

    • Haghverdi, Laleh, Maren Büttner, F. Alexander Wolf, Florian Buettner, and Fabian J. Theis. “Diffusion Pseudotime Robustly Reconstructs Lineage Branching.” Nature Methods 13, no. 10 (2016): 845–48. https://doi.org/10.1038/nmeth.3971.

  • TSCAN - pseudo-time reconstruction for scRNA-seq. Clustering first, then minimum spanning tree over cluster centers. Cells are projected to the tree (PCA) to determine their pseudo-time and order. Code https://github.com/zji90/tscan and R package that includes GUI http://www.bioconductor.org/packages/release/bioc/html/TSCAN.html

    • Ji, Zhicheng, and Hongkai Ji. “TSCAN: Pseudo-Time Reconstruction and Evaluation in Single-Cell RNA-Seq Analysis.” Nucleic Acids Research 44, no. 13 (27 2016): e117. https://doi.org/10.1093/nar/gkw430.

  • Wishbone - ordering scRNA-seq along bifurcating developmental trajectories. nearest-heighbor graphs to capture developmental distances using shortest paths. Solves short-circuits by low-dimensional projection using diffusion maps. Waypoints as guides for building the trajectory. Detailed and comprehensive Methods description. Supersedes Wanderlust. Comparison with SCUBA, Monocle. https://github.com/ManuSetty/wishbone

    • Setty, Manu, Michelle D. Tadmor, Shlomit Reich-Zeliger, Omer Angel, Tomer Meir Salame, Pooja Kathail, Kristy Choi, Sean Bendall, Nir Friedman, and Dana Pe’er. “Wishbone Identifies Bifurcating Developmental Trajectories from Single-Cell Data.” Nature Biotechnology 34, no. 6 (2016): 637–45. https://doi.org/10.1038/nbt.3569.

  • cellTree - hierarchical tree inference and visualization. Latent Dirichlet Allocation (LDA). Cells are analogous to text documents, discretized gene expression levels replace word frequencies. The LDA model represents topic distribution for each cell, analogous to low-dimensional embedding of the data where similarity is measured with chi-square distance. Fast and precise. http://bioconductor.org/packages/release/bioc/html/cellTree.html

    • duVerle, David A., Sohiya Yotsukura, Seitaro Nomura, Hiroyuki Aburatani, and Koji Tsuda. “CellTree: An R/Bioconductor Package to Infer the Hierarchical Structure of Cell Populations from Single-Cell RNA-Seq Data.” BMC Bioinformatics 17, no. 1 (December 2016). https://doi.org/10.1186/s12859-016-1175-6.

  • Monocle - Temporal ordering of single cell gene expression profiles. Independent Component Analysis to reduce dimensionality, Minimum Spanning Tree on the reduced representation and the longest path through it. https://cole-trapnell-lab.github.io/monocle-release/

    • Trapnell, Cole, Davide Cacchiarelli, Jonna Grimsby, Prapti Pokharel, Shuqiang Li, Michael Morse, Niall J. Lennon, Kenneth J. Livak, Tarjei S. Mikkelsen, and John L. Rinn. “The Dynamics and Regulators of Cell Fate Decisions Are Revealed by Pseudotemporal Ordering of Single Cells.” Nature Biotechnology 32, no. 4 (April 2014): 381–86. https://doi.org/10.1038/nbt.2859.

  • SCUBA - single-cell clustering using bifurcation analysis. Cells may differentiate in a monolineage manner or may differentiate into multiple cell lineages, which is the bifurcation event - two new lineages. Methods. Matlab code https://github.com/gcyuan/SCUBA

    • Marco, Eugenio, Robert L. Karp, Guoji Guo, Paul Robson, Adam H. Hart, Lorenzo Trippa, and Guo-Cheng Yuan. “Bifurcation Analysis of Single-Cell Gene Expression Data Reveals Epigenetic Landscape.” Proceedings of the National Academy of Sciences of the United States of America 111, no. 52 (December 30, 2014): E5643-5650. https://doi.org/10.1073/pnas.1408993111.

Networks

  • GraphDDP - combines user-guided clustering and transition of differentiation processes between clusters. Shortcomings of PCA, MDS, t-SNE. Tested on several datasets to improve interpretability of clustering, compared with other methods (Monocle2, SPRING, TSCAN). Detailed methods. https://github.com/fabriziocosta/GraphEmbed

    • Costa, Fabrizio, Dominic Grün, and Rolf Backofen. “GraphDDP: A Graph-Embedding Approach to Detect Differentiation Pathways in Single-Cell-Data Using Prior Class Knowledge.” Nature Communications 9, no. 1 (December 2018). https://doi.org/10.1038/s41467-018-05988-7.

  • PAGA - graph-like representation of scRNA-seq data. The kNN graph is partitioned using Louvain community detection algorithm, discarding spurious edged (denoising). Much faster than UMAP. Part of Scanpy pipeline.https://github.com/theislab/paga

    • Wolf, F. Alexander, Fiona K. Hamey, Mireya Plass, Jordi Solana, Joakim S. Dahlin, Berthold Göttgens, Nikolaus Rajewsky, Lukas Simon, and Fabian J. Theis. “PAGA: Graph Abstraction Reconciles Clustering with Trajectory Inference through a Topology Preserving Map of Single Cells.” Genome Biology 20, no. 1 (March 19, 2019): 59. https://doi.org/10.1186/s13059-019-1663-x.

  • SCENIC - single-cell network reconstruction and cell-state identification. Three modules: 1) GENIE3 - connect co-expressed genes and TFs using random forest regression; 2) RcisTarget - Refine them using cis-motif enrichment; 3) AUCell - assign activity scores for each network in each cell type. The R implementation of GENIE3 does not scale well with larger datasets; use arboreto instead, which is a much faster Python implementation. https://gbiomed.kuleuven.be/english/research/50000622/lcb/tools/scenic, https://github.com/aertslab/SCENIC, https://github.com/aertslab/GENIE3, https://github.com/aertslab/AUCell, https://github.com/tmoerman/arboreto, https://github.com/aertslab/pySCENIC

    • Aibar, Sara, Carmen Bravo González-Blas, Thomas Moerman, Vân Anh Huynh-Thu, Hana Imrichova, Gert Hulselmans, Florian Rambow, et al. “SCENIC: Single-Cell Regulatory Network Inference and Clustering.” Nature Methods 14, no. 11 (November 2017): 1083–86. https://doi.org/10.1038/nmeth.4463.
    • Davie et al. "A Single-Cell Transcriptome Atlas of the Aging Drosophila Brain" Cell, 2018 https://doi.org/10.1016/j.cell.2018.05.057

  • SCIRA - infer tissue-specific regulatory networks using large-scale bulk RNA-seq, estimate regulatory activity. SEPIRA uses a greedy partial correlation framework to infer a regulatory network from GTeX data, TF-specific regulons used as target profiles in a linear regression model framework. Compared against SCENIC. Works even for small cell populations. Tested on three scRNA-seq datasets. A part of SEPIRA R package, http://bioconductor.org/packages/release/bioc/html/SEPIRA.html

    • Wang, Ning, and Andrew E Teschendorff. “Leveraging High-Powered RNA-Seq Datasets to Improve Inference of Regulatory Activity in Single-Cell RNA-Seq Data.” BioRxiv, February 22, 2019. https://doi.org/10.1101/553040.

  • SINCERA - identification of major cell types, the corresponding gene signatures and transcription factor networks. Pre-filtering (expression filter, cell specificity filter) improves inter-sample correlation and decrease inter-sample distance. Normalization: per-sample z-score, then trimmed mean across cells. Clustering (centered Pearson for distance, average linkage), and other metrics, permutation to assess clustering significance. Functional enrichment, cell type enrichment analysis, identification of cell signatures. TF networks and their parameters (disruptive fragmentation centrality, disruptive connection centrality, disruptive distance centrality). Example analysis of mouse lung cells at E16.5, Fluidigm, 9 clusters, comparison with SNN-Cliq, scLVM, SINGuLAR Analysis Toolset. Web-site: https://research.cchmc.org/pbge/sincera.html; GitHub: https://github.com/xu-lab/SINCERA; Data

    • Guo, Minzhe, Hui Wang, S. Steven Potter, Jeffrey A. Whitsett, and Yan Xu. “SINCERA: A Pipeline for Single-Cell RNA-Seq Profiling Analysis.” PLoS Computational Biology 11, no. 11 (November 2015): e1004575. https://doi.org/10.1371/journal.pcbi.1004575.

RNA velocity

  • scVelo - more precise estimation of RNA velocity by solving the full transcriptional dynamics of splicing kinetics using a likelihood-based dynamical model. Description of steady-state (original), dynamical, and stochastic models. Ten-fold faster. https://scvelo.readthedocs.io/

    • Bergen, Volker, Marius Lange, Stefan Peidli, F. Alexander Wolf, and Fabian J. Theis. “Generalizing RNA Velocity to Transient Cell States through Dynamical Modeling.” Preprint. Bioinformatics, October 29, 2019. https://doi.org/10.1101/820936.

  • velocyto - RNA velocity, the time derivative of the gene expression state, estimated by the balance of spliced and unspliced mRNAs, and the mRNA degradation, in scRNA-seq (10X, inDrop, SMART-seq2, STRT/C1 protocols). Demonstrated on several datasets. Brief overview https://youtu.be/EPTgF4EA2zY. Python and R implementation http://velocyto.org/

    • La Manno, Gioele, Ruslan Soldatov, Amit Zeisel, Emelie Braun, Hannah Hochgerner, Viktor Petukhov, Katja Lidschreiber, et al. “RNA Velocity of Single Cells.” Nature 560, no. 7719 (August 2018): 494–98. https://doi.org/10.1038/s41586-018-0414-6.

  • RNA velocity in bulk RNA-seq data, https://github.com/praneet1988/Inferring-and-Visualizing-RNA-Velocity-in-Bulk-RNA-SEQ

Differential expression

  • Comparison of 11 differential gene expression detection methods for scRNA-seq data. Variable performance, poor overlap. Brief description of the statistics of each method. Bulk RNA-sec methods perform well, edgeR is good and fast. Table 1. Software tools for identifying DE genes using scRNA-seq data

    • Wang, Tianyu, Boyang Li, Craig E. Nelson, and Sheida Nabavi. “Comparative Analysis of Differential Gene Expression Analysis Tools for Single-Cell RNA Sequencing Data.” BMC Bioinformatics 20, no. 1 (December 2019): 40. https://doi.org/10.1186/s12859-019-2599-6.

  • DEsingle - an R package for detecting three types of differentially expressed genes in scRNA-seq data. Using Zero Inflated Negative Binomial distribution to distinguish true zeros from dropouts. Differential expression status (difference in the proportion of real zeros), DE abundance (typical DE without proportion of zeros change), DE general (both DE and proportion of zeros change). The majority of information is in supplementary material. https://bioconductor.org/packages/release/bioc/html/DEsingle.html

  • Soneson, Charlotte, and Mark D Robinson. “Bias, Robustness and Scalability in Single-Cell Differential Expression Analysis.” Nature Methods, February 26, 2018. https://doi.org/10.1038/nmeth.4612. - Differential analysis of scRNA-seq data, 36 methods. Prefiltering of low expressed genes is important. edgeRQLFDetRate performs best. The conquer database, scRNA-seq datasets as RDS objects - http://imlspenticton.uzh.ch:3838/conquer/

  • scDD R package to identify differentially expressed genes in single cell RNA-seq data. Accounts for unobserved data. Four types of differential expression (DE, DP, DM, DB, see paper). https://github.com/kdkorthauer/scDD

    • Korthauer, Keegan D., Li-Fang Chu, Michael A. Newton, Yuan Li, James Thomson, Ron Stewart, and Christina Kendziorski. “A Statistical Approach for Identifying Differential Distributions in Single-Cell RNA-Seq Experiments.” Genome Biology 17, no. 1 (December 2016). https://doi.org/10.1186/s13059-016-1077-y.

  • MAST - scRNA-seq DEG analysis. CDR - the fraction of genes that are detectably expressed in each cell - added to the hurdle model that explicitly parameterizes distributions of expressed and non-expressed genes. Generalized linear model, log2(TPM+1), Gaussian. Regression coeffs are estimated using Bayesian approach. Variance shrinkage, gamma distribution. https://github.com/RGLab/MAST

    • Finak, Greg, Andrew McDavid, Masanao Yajima, Jingyuan Deng, Vivian Gersuk, Alex K. Shalek, Chloe K. Slichter, et al. “MAST: A Flexible Statistical Framework for Assessing Transcriptional Changes and Characterizing Heterogeneity in Single-Cell RNA Sequencing Data.” Genome Biology 16 (December 10, 2015): 278. https://doi.org/10.1186/s13059-015-0844-5.

  • SCDE. Stochasticity of gene expression, high drop-out rate. A mixture model of two processes - detected expression and drop-out failure modeled as low-magnitude Poisson. Drop-out rate depends on the expected expression and can be approximated by logistic regression. https://hms-dbmi.github.io/scde/index.html

    • Kharchenko, Peter V., Lev Silberstein, and David T. Scadden. “Bayesian Approach to Single-Cell Differential Expression Analysis.” Nature Methods 11, no. 7 (July 2014): 740–42. https://doi.org/10.1038/nmeth.2967.

CNV

  • CaSpER - identification of CNVs from  RNA-seq data, bulk and single-cell (full-transcript only, like SMART-seq). Utilized multi-scale smoothed global gene expression profile and B-allele frequency (BAF) signal profile, detects concordant shifts in signal using a 5-state HMM (homozygous deletion, heterozygous deletion, neutral, one-copy-amplification, high-copy-amplification). Reconstructs subclonal CNV architecture for scRNA-seq data. Tested on GBM scRNA-seq, TCGA, other. Compared with HoneyBADGER. R code and tutorials https://github.com/akdess/CaSpER

    • Serin Harmanci, Akdes, Arif O. Harmanci, and Xiaobo Zhou. “CaSpER Identifies and Visualizes CNV Events by Integrative Analysis of Single-Cell or Bulk RNA-Sequencing Data.” Nature Communications 11, no. 1 (December 2020): 89. https://doi.org/10.1038/s41467-019-13779-x.

  • CNV estimation algorithm in scRNA-seq data - moving 100-gene window, deviation of expression from the chromosome average. Details in Methods.

    • Tirosh, Itay, Andrew S. Venteicher, Christine Hebert, Leah E. Escalante, Anoop P. Patel, Keren Yizhak, Jonathan M. Fisher, et al. “Single-Cell RNA-Seq Supports a Developmental Hierarchy in Human Oligodendroglioma.” Nature 539, no. 7628 (November 2016): 309–13. https://doi.org/10.1038/nature20123.

Annotation, subpopulation identification

  • Garnett - annotating cells in scRNA-seq data. Hierarchy of cell types and their markers should be pre-defined using a markup language. A classifier is trained to classify additional datasets. Trained on cells from one organisms, can be applied to different organisms. Pre-trained classifiers available. R-based. https://cole-trapnell-lab.github.io/garnett/

    • Pliner, Hannah A., Jay Shendure, and Cole Trapnell. “Supervised Classification Enables Rapid Annotation of Cell Atlases.” Nature Methods 16, no. 10 (October 2019): 983–86. https://doi.org/10.1038/s41592-019-0535-3

  • scPopCorn - subpopulation identification across scRNA-seq experiments. Identifies shared and unique subpopulations. Joint network of two graphs. First, graphs are built for each experiment using co-expression to identify subpopulations. Second, the corresponsence of the identified subpopulations is refined using Google's PageRank algorithm to identify subpopulations. Compared with Seurat alignment + Louvain, mutual nearest neighbor (MNN) method, and MNN + Louvain. Several assessment metrics. Tested on pancreatic, kidney cells, healthy brain and glioblastoma scRNA-seq data. Sankey diagrams showing how subpopulation assignment change. https://github.com/ncbi/scPopCorn/

    • Wang, Yijie, Jan Hoinka, and Teresa M. Przytycka. “Subpopulation Detection and Their Comparative Analysis across Single-Cell Experiments with ScPopCorn.” Cell Systems 8, no. 6 (June 2019): 506-513.e5. https://doi.org/10.1016/j.cels.2019.05.007.

  • SingleR - scRNA-seq cell type assignment by (Spearman) correlating to reference bulk RNA-seq data of pure cell types. Validated on ImmGen data. The package provides Human Primary Cell Atlas data, Blueprint and ENCODE consortium data, ImmGen, three others as a reference. Post-Seurat analysis. Web tool, http://comphealth.ucsf.edu/SingleR/, that takes SingleR objects, instructions are on GitHub, https://github.com/dviraran/SingleR/. Example analysis: http://comphealth.ucsf.edu/sample-apps/SingleR/SingleR.MCA.html, Bioconductor package https://bioconductor.org/packages/devel/bioc/html/SingleR.html, Twitter

    • Aran, Dvir, Agnieszka P. Looney, Leqian Liu, Esther Wu, Valerie Fong, Austin Hsu, Suzanna Chak, et al. “Reference-Based Analysis of Lung Single-Cell Sequencing Reveals a Transitional Profibrotic Macrophage.” Nature Immunology 20, no. 2 (February 2019): 163–72. https://doi.org/10.1038/s41590-018-0276-y.

  • TooManyCells - divisive hierarchical spectral clustering of scRNA-seq data. Uses truncated singular vector decomposition to bipartition the cells. Newman-Girvain modularity Q to assess whether bipartition is significant or should be stopped. BirchBeer visualization. Outperforms Phenograph, Seurat, Cellranger, Monocle, the latter is second in performance. Excels for rare populations. Normalization marginally affects performance. https://github.com/GregorySchwartz/tooManyCellsR and  https://github.com/faryabiLab/birch-beer

  • SingleCellNet - quantitative cell type annotation. Top-scoring pair transformation to match query and reference datasets. Compared with SCMAP, binary cell type classifier based on correlation. Benchmarked on 12 scRNA-seq datasets, provided in the GitHub repo, http://github.com/pcahan1/singleCellNet/. Blog post

    • Tan, Yuqi, and Patrick Cahan. “SingleCellNet: A Computational Tool to Classify Single Cell RNA-Seq Data Across Platforms and Across Species.” Cell Systems, July 2019. https://doi.org/10.1016/j.cels.2019.06.004.

  • CellAssign - R package for scRNA-seq cell type inference. Probabilistic graphical model to assign cell type probabilities to single cells using known marker genes (binarized matrix), including "unassigned" categorization. Insensitive to batch- or sample-specific effects. Outperforms Seurat, SC3, PhenoGraph, densityCut, dynamicTreeCut, scmap-cluster, correlation-based methods, SCINA. Applied to delineate the composition of the tumor microenvironment. Built using TensorFlow. https://github.com/irrationone/cellassign

    • Zhang, Allen W., Ciara O’Flanagan, Elizabeth A. Chavez, Jamie L. P. Lim, Nicholas Ceglia, Andrew McPherson, Matt Wiens, et al. “Probabilistic Cell-Type Assignment of Single-Cell RNA-Seq for Tumor Microenvironment Profiling.” Nature Methods, September 9, 2019. https://doi.org/10.1038/s41592-019-0529-1.

  • matchSCore2 - classifying cell types based on reference data. https://github.com/elimereu/matchSCore2

    • Mereu, Elisabetta, Atefeh Lafzi, Catia Moutinho, Christoph Ziegenhain, Davis J. MacCarthy, Adrian Alvarez, Eduard Batlle, et al. “Benchmarking Single-Cell RNA Sequencing Protocols for Cell Atlas Projects.” Preprint. Genomics, May 13, 2019. https://doi.org/10.1101/630087.

  • Single-Cell Signature Explorer - gene signature (~17,000 from MSigDb, KEGG, Reactome) scoring (sum of UMIs in in a gene signature over the total UMIs in a cell) for single cells, and visualization on top of a t-SNE plot. Optional Noise Reduction (Freeman-Tuckey transform to stabilize technical noise). Four consecutive tools (Go language, R/Shiny). Comparison with Seurat's Cell CycleScore module and AUCell from SCENIC. Very fast.https://sites.google.com/site/fredsoftwares/products/single-cell-signature-explorer

    • Pont, Frédéric, Marie Tosolini, and Jean Jacques Fournié. “Single-Cell Signature Explorer for Comprehensive Visualization of Single Cell Signatures across ScRNA-Seq Data Sets.” Preprint. Bioinformatics, April 29, 2019. https://doi.org/10.1101/621805.

  • VISION - functional annotation of scRNA-seq data using gene signatures (Geary's C statistics), unsupervised and supervised. Operates downstream of dimensionality reduction, clustering. A continuation of FastProject. https://github.com/YosefLab/VISION

    • DeTomaso, David, Matthew Jones, Meena Subramaniam, Tal Ashuach, Chun J Ye, and Nir Yosef. “Functional Interpretation of Single-Cell Similarity Maps,” August 29, 2018. https://doi.org/10.1101/403055.

Cell markers

Simulation

  • scDesign - scRNA-seq data simulator and statistical framework to access experimental design for differential gene expression analysis. Gamma-Normal mixture model better fits scRNA-seq data, accounts for dropout events (Methods describe step-wise statistical derivations). Single- or double-batch sequencing scenarios. Comparable or superior performance to simulation methods splat, powsimR, scDD, Lun et al. method. DE tested using t-test. Applications include DE methods evaluation, dimensionality reduction testing. https://github.com/Vivianstats/scDesign

  • Splatter - scRNA-seq simulator and pre-defined differential expression. 6 methods, description of each. Issues with scRNA-seq data - dropouts, zero inflation, proportion of zeros, batch effect. Negative binomial for simulation. No simulation is perfect. https://github.com/Oshlack/splatter

Power

  • How many cells do we need to sample so that we see at least n cells of each type? https://satijalab.org/howmanycells

  • scPower - an R package for power calculation for single-cell RNA-seq studies. Estimates power of differential expression and eQTLs using zero-inflated negative binomial distribution. Also, power to detect rare cell types. Figure 1 shows the dependence among experimental design parameters. Tested on several datasets, generalizes well across technologies. GitHub https://github.com/heiniglab/scPower and Shiny app http://scpower.helmholtz-muenchen.de/

  • powsimR - an R package for simulating scRNA-seq datasets and assess performance of differential analysis methods. Supports Poisson, Negative Binomial, and zero inflated NB, or estimates parameters from user-provided data. Simulates differential expression with pre-defined fold changes, estimates power, TPR, FDR, sample size, and for the user-provided dataset. https://github.com/bvieth/powsimR

    • Vieth, Beate, Christoph Ziegenhain, Swati Parekh, Wolfgang Enard, and Ines Hellmann. “PowsimR: Power Analysis for Bulk and Single Cell RNA-Seq Experiments.” Edited by Ivo Hofacker. Bioinformatics 33, no. 21 (November 1, 2017): 3486–88. https://doi.org/10.1093/bioinformatics/btx435.

Benchmarking

  • CellBench - an R package for benchmarking of scRNA-seq analysis pipelines. Simulated datasets using mixtures of either cells of RNA from five cancer cell lines, dilution series, ERCC spike-in controls. Four technologies. Methods: normalization, imputation, clustering, trajectory analysis, data integration. Evaluation metrics: silhouette width, correlations, others. Best performers: Normalization - Linnorm, scran, scone; Imputation - kNN, DrImpute; Clustering - all methods are OK, Seurat performs well; Trajectory - Slingshot and Monocle2. Processed datasets used for the analysis, https://github.com/LuyiTian/sc_mixology, R package, https://github.com/Shians/CellBench, https://bioconductor.org/packages/release/bioc/html/CellBench.html
    • Tian, Luyi, Xueyi Dong, Saskia Freytag, Kim-Anh Lê Cao, Shian Su, Abolfazl JalalAbadi, Daniela Amann-Zalcenstein, et al. “Benchmarking Single Cell RNA-Sequencing Analysis Pipelines Using Mixture Control Experiments.” Nature Methods, May 27, 2019. https://doi.org/10.1038/s41592-019-0425-8.

Deep learning

  • SAVER-X - denoising scRNA-seq data using deep autoencoder with a Bayesian model. Decomposes the variation into three components: 1) predictable, 2) unpredictable, 3) technical noise. Pretrained on the Human Cell Atlas project, 10X Genomics immune cells, allows for human-mouse cross-species learning. Improves clustering and the detection of differential genes. Outperforms downsampling, MAGIC, DCA, scImpute. https://github.com/jingshuw/SAVERX

    • Littmann, Maria, Katharina Selig, Liel Cohen-Lavi, Yotam Frank, Peter Hönigschmid, Evans Kataka, Anja Mösch, et al. “Validity of Machine Learning in Biology and Medicine Increased through Collaborations across Fields of Expertise.” Nature Machine Intelligence, January 13, 2020. https://doi.org/10.1038/s42256-019-0139-8.

  • scVI - low-dimensional representation of scRNA-seq data used for batch correction, imputation, clustering, differential expression. Deep neural networks to approximate the distribution that underlie observed expression values. Zero-inflated negative binomial distribution conditioned on the batch annotation and unobserved random variables. Compared with DCA, ZINB-WAVE on simulated and real large and small datasets. Perspective by Way & Greene https://github.com/YosefLab/scVI

    • Lopez, Romain, Jeffrey Regier, Michael B Cole, Michael Jordan, and Nir Yosef. “Bayesian Inference for a Generative Model of Transcriptome Profiles from Single-Cell RNA Sequencing,” September 23, 2018. https://doi.org/10.1101/292037.

Spatial transcriptomics

scATAC-seq integration, Multi-omics methods

  • Multi-omics methods - Table 1 from Sierant, Michael C., and Jungmin Choi. “Single-Cell Sequencing in Cancer: Recent Applications to Immunogenomics and Multi-Omics Tools.” Genomics & Informatics 16, no. 4 (December 2018): e17. https://doi.org/10.5808/GI.2018.16.4.e17.

  • scAI - integrative analysis of scRNA-seq and scATAC-seq or scMethylation data measured from the same cells (in contrast to different measures sampled from the same cell population). Overview of multi-omics single-cell technologies, methods for data integration in bulk samples and single-cell samples (MATCHER, Seural, LIGER), sparsity (scATAC-seq is ~99% sparse and nearly binary). Deconvolution of both single-cell matrices into gene loading and locus loading matrices, a cell loading matrix, in which factors K correspond to loadings of gene, locus, and cell in the K-dimensional space. A strategy to reduce over-aggregation. Cell subpopulations identified by Leiden clustering of the cell loading matrix. Visualization of the low-rank matrices with the Sammon mapping. Multi-omics simulation using MOSim, eight scenarios of simulated data, AUROC and Normalized Mutual Information (NMI) assessment of matrix reconstruction quality. Compared with MOFA, Seurat, LIGER. Tested on 8837 mammalian kidney cells scRNA-seq and scATAC-seq data, 77 mouse ESCs scRNA-seq and scMethylation, interpretation. https://github.com/sqjin/scAI

    • Jin, Suoqin, Lihua Zhang, and Qing Nie. “ScAI: An Unsupervised Approach for the Integrative Analysis of Parallel Single-Cell Transcriptomic and Epigenomic Profiles.” Genome Biology 21, no. 1 (December 2020): 25. https://doi.org/10.1186/s13059-020-1932-8.

  • Harmony - scRNA-seq integration by projecting datasets into a shared embedding where cells differences between cell clusters are maximized while differences between datasets of origin are minimized = focus on clusters driven by cell differences. Can account for technical and biological covariates. Can integrate scRNA-seq datasets obtained with different technologies, or scRNA- and scATAC-seq, scRNA-seq with spatially-resolved transcriptomics. Local inverse Simpson Index (LISI) to test for database- and cell-type-specifc clustering. Outperforms MNN, BBKNN, MultiCCA, Scanorama. Memory-efficient, fast, scales to large datasets, included in Seurat. https://github.com/immunogenomics/harmony, https://github.com/slowkow/harmonypy

    • Korsunsky, Ilya, Nghia Millard, Jean Fan, Kamil Slowikowski, Fan Zhang, Kevin Wei, Yuriy Baglaenko, Michael Brenner, Po-ru Loh, and Soumya Raychaudhuri. “Fast, Sensitive and Accurate Integration of Single-Cell Data with Harmony.” Nature Methods 16, no. 12 (December 2019): 1289–96. https://doi.org/10.1038/s41592-019-0619-0.

  • Seurat v.3 paper. Integration of multiple scRNA-seq and other single-cell omics (spatial transcriptomics, scATAC-seq, immunophenotyping), including batch correction. Anchors as reference to harmonize multiple datasets. Canonical Correlation Analysis (CCA) coupled with Munual Nearest Neighborhoors (MNN) to identify shared subpopulations across datasets. CCA to reduce dimensionality, search for MNN in the low-dimensional representation. Shared Nearest Neighbor (SNN) graphs to assess similarity between two cells. Outperforms scmap. Extensive validation on multiple datasets (Human Cell Atlas, STARmap mouse visual cortex spatial transcriptomics. Tabula Muris, 10X Genomics datasets, others in STAR methods). Data normalization, variable feature selection within- and between datasets, anchor identification using CCA (methods), their scoring, batch correction, label transfer, imputation. Methods correspond to details of each Seurat function. Preprocessing of real single-cell data.https://satijalab.org/seurat/, https://github.com/satijalab/Integration2019

  • Signac is an extension of Seurat for the analysis, interpretation, and exploration of single-cell chromatin datasets. https://satijalab.org/signac/

    • Stuart, Tim, Andrew Butler, Paul Hoffman, Christoph Hafemeister, Efthymia Papalexi, William M Mauck, Marlon Stoeckius, Peter Smibert, and Rahul Satija. “Comprehensive Integration of Single Cell Data.” Preprint. Genomics, November 2, 2018. https://doi.org/10.1101/460147.

  • MAESTRO - MAESTRO (Model-based AnalysEs of Single-cell Transcriptome and RegulOme) is a comprehensive single-cell RNA-seq and ATAC-seq analysis suit built using snakemake. https://github.com/liulab-dfci/MAESTRO, Tweet

10X Genomics

10X QC

Data

Human

Cancer

Mouse

  • scRNA-seq of mouse gastrulation and early embryogenesis.  116,312 cells, nine time points from 6.5 to 8.5 days post-fertilization. Methods, scran-based analysis Online version with tutorial https://marionilab.cruk.cam.ac.uk/MouseGastrulation2018/, download script at https://github.com/MarioniLab/EmbryoTimecourse2018, Bioconductor data package https://github.com/MarioniLab/MouseGastrulationData

    • Pijuan-Sala, Blanca, Jonathan A. Griffiths, Carolina Guibentif, Tom W. Hiscock, Wajid Jawaid, Fernando J. Calero-Nieto, Carla Mulas, et al. “A Single-Cell Molecular Map of Mouse Gastrulation and Early Organogenesis.” Nature, February 20, 2019. https://doi.org/10.1038/s41586-019-0933-9.

  • DropViz - Exploring the Mouse Brain through Single Cell Expression Profiles. Drop-seq to analyze 690,000 individual cells from nine different regions of the adult mouse brain. http://dropviz.org/

  • STARmap - in situ gene expression datasets of the mouse visual cortex (890 cells, 1,020 genes). https://www.starmapresources.com/data/

  • Mouse Cell Atlas - http://bis.zju.edu.cn/MCA/

  • Tablua Muris - Mouse scRNA-seq of 100,605 cells from 20 organs and tissues. Web-site http://tabula-muris.ds.czbiohub.org/, and GitHub with gene matrix download and other scripts to reproduce all figures in the paper, https://github.com/czbiohub/tabula-muris. Raw data are at https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE109774. Processed: https://figshare.com/articles/Single-cell_RNA-seq_data_from_Smart-seq2_sequencing_of_FACS_sorted_cells_v2_/5829687/1

    • Quake, Stephen R., Tony Wyss-Coray, and Spyros Darmanis. “Single-Cell Transcriptomic Characterization of 20 Organs and Tissues from Individual Mice Creates a Tabula Muris,” March 29, 2018. https://doi.org/10.1101/237446.

  • The 1 million neuron data set from E18 mice, from the 10X Genomics website (https://support.10xgenomics.com/single-cell-gene-expression/datasets). R packages for its analyses:

  • Single-cell ATAC-seq, approx. 100,000 single cells from 13 adult mouse tissues. Two sequence platforms, good concordance. Filtered data assigned into 85 clusters. Genes associated with the corresponding ATAC sites (Cicero for identification). Differential accessibility. Motif enrichment (Basset CNN). GWAS results enrichment. All data and metadata are available for download as text or rds format at http://atlas.gs.washington.edu/mouse-atac/

    • Cusanovich, Darren A., Andrew J. Hill, Delasa Aghamirzaie, Riza M. Daza, Hannah A. Pliner, Joel B. Berletch, Galina N. Filippova, et al. “A Single-Cell Atlas of In Vivo Mammalian Chromatin Accessibility.” Cell 174, no. 5 (August 2018): 1309-1324.e18. https://doi.org/10.1016/j.cell.2018.06.052.

  • scRNA-seq of aging. >50,000 cells from kidney, lung, and spleen in young (7 months) and aged (22-23 months) mice. Transcriptional variation using difference from the median. Cell-cell heterogeneity using the Euclidean distance from centroids. Aging trajectories derived from NMF embedding. Cell type identification by neural network trained on Tabula Muris. ln(CPM + 1) UMIs used for all analyses. Visualization, downloadable data, code, pre-trained network.  https://mca.research.calicolabs.com/

    • Kimmel, Jacob C, Lolita Penland, Nimrod D Rubinstein, David G Hendrickson, David R Kelley, and Adam Z Rosenthal. “A Murine Aging Cell Atlas Reveals Cell Identity and Tissue-Specific Trajectories of Aging.” BioRxiv, January 1, 2019, 657726. https://doi.org/10.1101/657726.

  • scRNA-seq (10X Genomics) of murine cerebellum. 39245 cells (after filtering) from 12 time points, 48 distinct clusters. Cell Seek for online exploration, https://cellseek.stjude.org/cerebellum/ and https://gawadlab.github.io/CellSeek/. Approx. 83,000 cells (BAM files per time point) are available at https://www.ebi.ac.uk/ena/data/view/PRJEB23051. No annotations.

    • Carter, Robert A., Laure Bihannic, Celeste Rosencrance, Jennifer L. Hadley, Yiai Tong, Timothy N. Phoenix, Sivaraman Natarajan, John Easton, Paul A. Northcott, and Charles Gawad. “A Single-Cell Transcriptional Atlas of the Developing Murine Cerebellum.” Current Biology, September 2018. https://doi.org/10.1016/j.cub.2018.07.062.

  • Sci-CAR, single-cell RNA- and ATAC-seq. Two experiments: 1) Lung adenocarcinoma A549 cells, dexametasone treatment over 3 timepoints. 2) Mixture of HEK293T (human) and NIH3T3 (mouse) cells. Differential gene expression, accessibility analysis, clustering. Linking distal open chromatin to genes, 44% map to nearest, 21 to the second nearest. Gene expression counts and ATAC-seq peaks, https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE117089

    • Cao, Junyue, Darren A. Cusanovich, Vijay Ramani, Delasa Aghamirzaie, Hannah A. Pliner, Andrew J. Hill, Riza M. Daza, et al. “Joint Profiling of Chromatin Accessibility and Gene Expression in Thousands of Single Cells.” Science 361, no. 6409 (September 28, 2018): 1380–85. https://doi.org/10.1126/science.aau0730.

  • Mouse spinal cord development scRNA-seq. Temporal (embryonic day 9.5-13.5) and spatial (cervical and thoracic regions of the neural tube) profiling. 10X genomics protocol, Cell Ranger processing, filtering, combinatorial testing for differential expression, pseudotime reconstruction using Monocle2. UMI matrix (21465 cells), https://www.ebi.ac.uk/arrayexpress/experiments/E-MTAB-7320/, code, https://github.com/juliendelile/MouseSpinalCordAtlas, Table S1 - binarized matrix of gene markers for each neuronal subtype, http://www.biologists.com/DEV_Movies/DEV173807/TableS1.csv

    • Delile, Julien, Teresa Rayon, Manuela Melchionda, Amelia Edwards, James Briscoe, and Andreas Sagner. “Single Cell Transcriptomics Reveals Spatial and Temporal Dynamics of Gene Expression in the Developing Mouse Spinal Cord.” Development, March 7, 2019, dev.173807. https://doi.org/10.1242/dev.173807.

  • Mouse pancreas scRNA-seq, ~12,000 cells. https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE84133. Deconvolution of bulk RNA-seq data using cell signatures derived from clusters of scRNA-seq cells, https://github.com/shenorrLab/bseqsc

    • Baron, Maayan, Adrian Veres, Samuel L. Wolock, Aubrey L. Faust, Renaud Gaujoux, Amedeo Vetere, Jennifer Hyoje Ryu, et al. “A Single-Cell Transcriptomic Map of the Human and Mouse Pancreas Reveals Inter- and Intra-Cell Population Structure.” Cell Systems 3, no. 4 (26 2016): 346-360.e4. https://doi.org/10.1016/j.cels.2016.08.011.

  • The full-length SMART-Seq2 protocol with deep sequencing (6,558 genes/cell) to profile 765 multipotent mouse hematopoietic progenitors. https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE81682

    • Nestorowa, S., F. K. Hamey, B. Pijuan Sala, E. Diamanti, M. Shepherd, E. Laurenti, N. K. Wilson, D. G. Kent, and B. Gottgens. “A Single-Cell Resolution Map of Mouse Hematopoietic Stem and Progenitor Cell Differentiation.” Blood 128, no. 8 (August 25, 2016): e20–31. https://doi.org/10.1182/blood-2016-05-716480.

  • Buettner, Florian, Kedar N Natarajan, F Paolo Casale, Valentina Proserpio, Antonio Scialdone, Fabian J Theis, Sarah A Teichmann, John C Marioni, and Oliver Stegle. “Computational Analysis of Cell-to-Cell Heterogeneity in Single-Cell RNA-Sequencing Data Reveals Hidden Subpopulations of Cells.” Nature Biotechnology 33, no. 2 (March 2015): 155–60. https://doi.org/10.1038/nbt.3102.

    • data/scLVM/nbt.3102-S7.xlsx - Uncorrected and cell-cycle corrected expression values (81 cells x 7073 genes) for T-cell data. Includes cluster assignment to naive T cells vs. TH2 cells (GATA3 high marker). Source
    • data/scLVM/nbt.3102-S8.xlsx - Corrected and uncorrected expression values for the newly generated mouse ESC data. 182 samples x 9571 genes. Source

  • Zeisel, A., Munoz-Manchado, A.B., Codeluppi, S., Lonnerberg, P., La Manno, G., Jureus, A., Marques, S., Munguba, H., He, L., Betsholtz, C., et al. (2015). Brain structure. Cell types in the mouse cortex and hippocampus revealed by single-cell RNA- seq. Science 347, 1138–1142. - 3,005 single cells from the hippocampus and cerebral cortex of mice. https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE60361, http://linnarssonlab.org/cortex/, and more on this site.

    • data/Brain/Zeisel_2015_TableS1.xlsx - Table S1 - gene signatures for Ependymal, Oligodendrocyte, Microglia, CA1 Pyramidal, Interneuron, Endothelial, S1 Pyramidal, Astrocyte, Mural cells. Source
    • data/Brain/expression_mRNA_17-Aug-2014.txt - 19,972 genes x 3005 cells. Additional rows with class annotations to interneurons, pyramidal SS, pyramidal CA1, oligodendrocytes, microglia, endothelial-mural, astrocytes_ependymal, further subdivided into 47 subclasses. Source

Brain single-cell data

  • scRNA-seq and scATAC-seq integration, human forebrain development timecourse. Analysis, reporting, data https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE132403

    • Trevino, Alexandro E., Nasa Sinnott-Armstrong, Jimena Andersen, Se-Jin Yoon, Nina Huber, Jonathan K. Pritchard, Howard Y. Chang, William J. Greenleaf, and Sergiu P. Pașca. “Chromatin Accessibility Dynamics in a Model of Human Forebrain Development.” Science 367, no. 6476 (January 24, 2020): eaay1645. https://doi.org/10.1126/science.aay1645.

  • scRNA-seq of the middle temporal gyrus of human cerebral cortex. Similarity with mouse cortex scRNA-seq, and differences (proportions, laminar distribution, gene expression, morphology). NeuN-labeled sorting to select for neuronal nuclei. SMART-Seq v4 library preparation. Downstream analysis includes PCA, clustering using Jaccard, Louvain community detection. 75 cell types. Gene expression matrix, clusters, more at http://celltypes.brain-map.org/rnaseq, interactive exploration (download) at https://viewer.cytosplore.org/

    • Hodge, Rebecca D, Trygve E Bakken, Jeremy A Miller, Kimberly A Smith, Eliza R Barkan, Lucas T Graybuck, Jennie L Close, et al. “Conserved Cell Types with Divergent Features between Human and Mouse Cortex.” Preprint. Neuroscience, August 5, 2018. https://doi.org/10.1101/384826.

  • Brain immune atlas scRNA-seq resource. Border-associated macrophages from discrete mouse brain compartments, tissue-specific transcriptional signatures. http://www.brainimmuneatlas.org/index.php, https://github.com/saeyslab/brainimmuneatlas/

    • Van Hove, Hannah, Liesbet Martens, Isabelle Scheyltjens, Karen De Vlaminck, Ana Rita Pombo Antunes, Sofie De Prijck, Niels Vandamme, et al. “A Single-Cell Atlas of Mouse Brain Macrophages Reveals Unique Transcriptional Identities Shaped by Ontogeny and Tissue Environment.” Nature Neuroscience, May 6, 2019. https://doi.org/10.1038/s41593-019-0393-4.

  • Darmanis, S., Sloan, S.A., Zhang, Y., Enge, M., Caneda, C., Shuer, L.M., Hayden Gephart, M.G., Barres, B.A., and Quake, S.R. (2015). A survey of human brain transcriptome diversity at the single cell level. Proc. Natl. Acad. Sci. USA 112, 7285–7290. - Single cell brain transcriptomics, human. Fluidigm C1 platform. Healthy cortex cells (466 cells) containing: Astrocytes, oligodendrocytes, oligodendrocyte precursor cells (OPCs), neurons, microglia, and vascular cells. Single cells clustered into 10 clusters, their top 20 gene signatures are in Supplementary Table S3. Raw data athttps://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE67835

    • data/Brain/TableS3.txt - top 20 cell type-specific genes
    • data/Brain/TableS3_matrix.txt - genes vs. cell types with 0/1 indicator variables.

  • Nowakowski, Tomasz J., Aparna Bhaduri, Alex A. Pollen, Beatriz Alvarado, Mohammed A. Mostajo-Radji, Elizabeth Di Lullo, Maximilian Haeussler, et al. “Spatiotemporal Gene Expression Trajectories Reveal Developmental Hierarchies of the Human Cortex.” Science 358, no. 6368 (December 8, 2017): 1318–23. https://doi.org/10.1126/science.aap8809. - single-cell RNA-seq of neuronal cell types. Dimensionality reduction, clustering, WGCNA, defining cell type-specific signatures, comparison with other signatures (Zeng, Miller).

    • data/Brain/Nowakowski_2017_Tables_S1-S11.xlsx - Table S5 has brain region-specific gene signatures. Source

  • Luo, Chongyuan, Christopher L. Keown, Laurie Kurihara, Jingtian Zhou, Yupeng He, Junhao Li, Rosa Castanon, et al. “Single-Cell Methylomes Identify Neuronal Subtypes and Regulatory Elements in Mammalian Cortex.” Science (New York, N.Y.) 357, no. 6351 (11 2017): 600–604. https://doi.org/10.1126/science.aan3351. - single-cell methylation of human and mouse neuronal cells. Marker genes with cell type-specific methylation profiles - Table S3, http://science.sciencemag.org/content/suppl/2017/08/09/357.6351.600.DC1

  • Major Depressive Disorder Working Group of the Psychiatric Genomics Consortium et al., “Genetic Identification of Brain Cell Types Underlying Schizophrenia,” Nature Genetics 50, no. 6 (June 2018): 825–33, https://doi.org/10.1038/s41588-018-0129-5. - Cell-type specificity of schizophrenia SNPs judged by enrichment in expressed genes. scRNA-seq custom data collection. Difference between schizophrenia and neurological disorders.

    • data/Brain_cell_type_gene_expression.xlsx - Supplementary Table 4 - Specificity values for Karolinska scRNA-seq superset. Specificity represents the proportion of the total expression of a gene found in one cell type as compared to that in all cell types (i.e., the mean expression in one cell type divided by the mean expression in all cell types). Gene X cell type matrix. Level 1 (core cell types) and level 2 (extended collection of cell types) data. Source

  • Nowakowski, Tomasz J., Aparna Bhaduri, Alex A. Pollen, Beatriz Alvarado, Mohammed A. Mostajo-Radji, Elizabeth Di Lullo, Maximilian Haeussler, et al. “Spatiotemporal Gene Expression Trajectories Reveal Developmental Hierarchies of the Human Cortex.” Science 358, no. 6368 (December 8, 2017): 1318–23. https://doi.org/10.1126/science.aap8809. - single-cell RNA-seq of human neuronal cell types. Dimensionality reduction, clustering, WGCNA, defining cell type-specific signatures, comparison with other signatures (Zeng, Miller). Supplementary material at http://science.sciencemag.org/content/suppl/2017/12/06/358.6368.1318.DC1 Table 5 has gene signatures.https://www.wired.com/story/neuroscientists-just-launched-an-atlas-of-the-developing-human-brain/. Controlled access data on dbGAP, https://www.ncbi.nlm.nih.gov/projects/gap/cgi-bin/study.cgi?study_id=phs000989.v3.p1, summarized matrix with annotations is available on https://cells.ucsc.edu/?ds=cortex-dev

Links

Papers

  • The single cell studies database, over 1000 studies. Main database, Tweet by Valentine Svensson

  • scATACdb - list of scATAC-seq studies, Google Sheet by Caleb Lareau

  • Journal club on single-cell multimodal data technology and analysis - Data science seminar led by Levi Waldron

  • scRNA-seq pipeline benchmarking, beta regression to explain the variance in pipeline performance. Best settings: Genome mapping using STAR + GENCODE annotation, imputation using scone or SAVER is optional, scran or SCnorm for normalization, any differential expression method, e.g., edgeR and limma-trend, works OK. Filtering has no effect on performance. The most significant effect on performance is from normalization and library preparation choices. Pipelines tested with powsimR package https://github.com/bvieth/powsimR, data at https://github.com/bvieth/scRNA-seq-pipelines

    • Vieth, Beate, Swati Parekh, Christoph Ziegenhain, Wolfgang Enard, and Ines Hellmann. “A Systematic Evaluation of Single Cell RNA-Seq Analysis Pipelines.” BioRxiv, March 19, 2019. https://doi.org/10.1101/583013.

  • All steps in scRNA-seq analysis, QC (count depth, number of genes, % mitochondrial), normalization (global, downsampling, nonlinear), data correction (batch, denoising, imputation), feature selection, dimensionality reduction (PCA, diffusion maps, tSNE, UMAP), visualization, clustering (k-means, graph/community detection), annotation, trajectory inference (PAGA, Monocle), differential analysis (DESeq2, EdgeR, MAST), gene regulatory networks. Description of the bigger picture at each step, latest tools, their brief description, references. R-based Scater as the full pipeline for QC and preprocessing, Seurat for downstream analysis, scanpy Python pipeline. Links and refs for tutorials. https://github.com/theislab/single-cell-tutorial

    • Luecken, Malte D., and Fabian J. Theis. “Current Best Practices in Single-Cell RNA-Seq Analysis: A Tutorial.” Molecular Systems Biology 15, no. 6 (June 19, 2019): e8746. https://doi.org/10.15252/msb.20188746.

  • Review of single-cell sequencing technologies, individual and combined, technical details of each. Combinatorial indexing. Genomic DNA, methylomes, histone modifications, open chromatin, 3D genomics, proteomics, spatial transcriptomics. Table 1 - multiomics technologies, summary. Areas of application, in cancer and cell atlases. Future development, e.g., single-cell metabolomics.

  • sciRNA-seq - single-cell combinatorial indexing RNA-seq technology and sequencing of C. elegans, ~49,000 cells, 27 cell types. Data and R code to download it at http://atlas.gs.washington.edu/hub/

    • Cao, Junyue, Jonathan S. Packer, Vijay Ramani, Darren A. Cusanovich, Chau Huynh, Riza Daza, Xiaojie Qiu, et al. “Comprehensive Single-Cell Transcriptional Profiling of a Multicellular Organism.” Science 357, no. 6352 (August 18, 2017): 661–67. https://doi.org/10.1126/science.aam8940.

About

A list of scRNA-seq analysis tools

Resources

Readme

License

MIT license
Activity

Stars

1 star

Watchers

1 watching

Forks

0 forks
Report repository

Releases

No releases published

Packages

No packages published

Languages

  • R 100.0%