Abstract:Background: While single cell RNA sequencing becomes a powerful technology, the presence of the large number of zero counts represents a challenge for both wet-lab processing and data analysis. Imputation of these dropouts can now be performed by three categories of algorithms: Model or smoothing, Matrix theory or Deep learning. However, two fundamental questions remain unsettled: (1) whether imputation should be performed; (2) which imputation algorithm to use with various downstream applications. Notably, imputation is not commonly used in real scRNA-seq applications because of their uncertain benefits, concerns about false inferences in downstream applications, and the lack of in-depth benchmark. Methods: Here, we performed two tasks. First, we developed an algorithm using adaptive low-rank full matrix factorization (afMF) based on a previous limited implementation confined to using low rank matrix decomposition (ALRA). Second, to evaluate the impact of various imputation algorithms on downstream analyses, a new benchmark framework incorporating commonly used downstream applications was developed. This benchmark framework put emphasis on real datasets which had ground truth or matched bulk data such that algorithm performance was compared to more convinced data rather than less realistic simulated parameters. Results: Our results indicated that afMF and ALRA (matrix based) provided good imputation and outperformed raw log-normalization in various downstream applications. afMF outperformed ALRA in several evaluations (cell-level differential expression analysis, GSEA, classification, biomarker prediction, clustering, SC-bulk profiling similarity). Besides, afMF ranked among the top levels in automatic cell type annotation, trajectory inference by DPT, and AUCell & SCENIC. Both showed acceptable scalability, while afMF had longer running time. MAGIC (smoothing based) and AutoClass (deep learning based) also performed well but may produce false positives. In contrast, more complicated methods (other deep learning or model based) were prone to overfitting and data distortion. We also found that certain downstream algorithms are not compatible with imputation, including trajectory inference with Slingshot and cell-cell communication. Prior imputation either showed no improvement or generated false positive findings with these downstream applications. Conclusions: We hope this in-depth evaluation and the algorithm developed in this study can enhance the selection of appropriate imputation algorithm for specific scRNA-seq downstream analyses.

Deep Imputation Bi-stochastic Graph Regularized Matrix Factorization for Clustering Single-cell RNA-sequencing Data

Detecting cell type from single cell RNA sequencing based on deep bi-stochastic graph regularized matrix factorization

A Deep Matrix Factorization Based Approach for Single-Cell RNA-seq Data Clustering.

Deep structural clustering for single-cell RNA-seq data jointly through autoencoder and graph neural network

SSNMDI: a novel joint learning model of semi-supervised non-negative matrix factorization and data imputation for clustering of single-cell RNA-seq data

DeepCI: a Deep Learning Based Clustering Method for Single Cell RNA-seq Data

Clustering single-cell RNA-seq data with a model-based deep learning approach

Single-cell RNA-seq Data Clustering by Deep Information Fusion.

Deep learning-based clustering method for single-cell RNA data

Single-Cell Transcriptome Data Clustering via Multinomial Modeling and Adaptive Fuzzy K-Means Algorithm

scCDCG: Efficient Deep Structural Clustering for single-cell RNA-seq via Deep Cut-informed Graph Embedding

Graph-Regularized Non-Negative Matrix Factorization for Single-Cell Clustering in scRNA-Seq Data

Scdssc: Deep Sparse Subspace Clustering for Scrna-Seq Data.

Dimensionality Reduction for Single Cell RNA Sequencing Data Using Constrained Robust Non-Negative Matrix Factorization

Unsupervised Cluster Analysis and Gene Marker Extraction of scRNA-seq Data Based On Non-Negative Matrix Factorization

A Fusion Learning Model Based on Deep Learning for Single-Cell RNA Sequencing Data Clustering

scASDC: Attention Enhanced Structural Deep Clustering for Single-cell RNA-seq Data

BANMF-S: a blockwise accelerated non-negative matrix factorization framework with structural network constraints for single cell imputation

Deep Learning for clustering single-cell RNA-seq Data

Imputation in Scrna-seq Data Using Supervised Deep Generative Networks

Low-Rank Full Matrix Factorization for dropout imputation in single cell RNA-seq and benchmarking with imputation algorithms for downstream applications