A Review of scRNA-seq Imputation Methods
Bibliographic record
Abstract
Single-cell RNA sequencing (scRNA-seq) has emerged as a powerful tool for profiling gene expression at the individual cell level, enabling the discovery of cellular heterogeneity that traditional bulk RNA sequencing cannot capture. However, technical limitations such as low transcript capture efficiency, amplification biases, and limited sequencing depth have led to pervasive dropout events, where true gene expression is obscured by excessive zero counts. This review systematically examines and compares the principal imputation methods developed to address these challenges in scRNA-seq data analysis. We categorize these approaches into two broad groups: model-based methods and deep learning methods. Model-based techniques utilize probabilistic models or matrix factorization to exploit similarities among cells and genes—either independently or in combination—to predict and restore missing values. In contrast, deep learning methods leverage the capabilities of autoencoders, graph neural networks, and other innovative network architectures, including generative adversarial networks, to capture complex nonlinear relationships within high-dimensional, noisy data. While model-based approaches offer greater interpretability through explicit statistical assumptions, they are often limited by their sensitivity to noise and data sparsity. Deep learning strategies, although computationally intensive and less interpretable, excel in recovering intricate data structures in large-scale datasets. By providing a comprehensive overview of these imputation strategies, this review aims to guide researchers in selecting the most appropriate methods for their specific datasets and downstream analyses, and to suggest future directions for improving imputation accuracy and integrating multi-omics data.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.008 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.003 | 0.005 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.003 | 0.001 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.005 | 0.004 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".