Abstract:Single-cell RNA sequencing (scRNA-seq) has revolutionized our understanding of cellular heterogeneity and tissue transcriptomic complexity. However, the high frequency of dropout events in scRNA-seq data complicates downstream analyses such as cell type identification and trajectory inference. Existing imputation methods address the dropout problem but face limitations such as high computational cost and risk of over-imputation. We present SmartImpute, a novel computational framework designed for targeted imputation of scRNA-seq data. SmartImpute focuses on a predefined set of marker genes, enhancing the biological relevance and computational efficiency of the imputation process while minimizing the risk of model misspecification. Utilizing a modified Generative Adversarial Imputation Network architecture, SmartImpute accurately imputes the missing gene expression and distinguishes between true biological zeros and missing values, preventing overfitting and preserving biologically relevant zeros. To ensure reproducibility, we also provide a function based on the GPT4 model to create target gene panels depending on the tissue types and research context. Our results, based on scRNA-seq data from head and neck squamous cell carcinoma and human bone marrow, demonstrate that SmartImpute significantly enhances cell type annotation and clustering accuracy while reducing computational burden. Benchmarking against other imputation methods highlights SmartImpute's superior performance in terms of both accuracy and efficiency. Overall, SmartImpute provides a lightweight, efficient, and biologically relevant solution for addressing dropout events in scRNA-seq data, facilitating deeper insights into cellular heterogeneity and disease progression. Furthermore, SmartImpute's targeted approach can be extended to spatial omics data, which also contain many missing values.

EnImpute: imputing dropout events in single-cell RNA-sequencing data via ensemble learning

A Posterior Probability Based Bayesian Method for Single-Cell RNA-seq Data Imputation.

AGImpute: imputation of scRNA-seq data based on a hybrid GAN with dropouts identification

SAE-Impute: imputation for single-cell data via subspace regression and auto-encoders

CL-Impute: A Contrastive Learning-Based Imputation for Dropout Single-Cell RNA-seq Data.

CDSImpute: an Ensemble Similarity Imputation Method for Single-Cell RNA Sequence Dropouts

Imputing dropouts for single-cell RNA sequencing based on multi-objective optimization

Schinter: Imputing Dropout Events for Single-Cell RNA-seq Data with Limited Sample Size

Regulatory Network-Based Imputation of Dropouts in Single-Cell RNA Sequencing Data

scNMF-Impute: imputation for single-cell RNA-seq data based on nonnegative matrix factorization.

GE-Impute: graph embedding-based imputation for single-cell RNA-seq data

scCGImpute: An Imputation Method for Single-Cell RNA Sequencing Data Based on Similarities between Cells and Relationships among Genes

CMF-Impute: an Accurate Imputation Tool for Single-Cell RNA-seq Data.

DeepImpute: an Accurate, Fast, and Scalable Deep Neural Network Method to Impute Single-Cell RNA-seq Data

scRMD: Imputation for single cell RNA-seq data via robust matrix decomposition

ccImpute: an accurate and scalable consensus clustering based algorithm to impute dropout events in the single-cell RNA-seq data

SCC: an accurate imputation method for scRNA-seq dropouts based on a mixture model

Imputation method for single-cell RNA-seq data using neural topic model

SmartImpute: A Targeted Imputation Framework for Single-cell Transcriptome Data

Imputing single-cell RNA-seq data by considering cell heterogeneity and prior expression of dropouts

AdImpute: An Imputation Method for Single-Cell RNA-Seq Data Based on Semi-Supervised Autoencoders