Table 1.
Comparative overview of some DEA methods for scRNA-seq data.
Fig 1.
Research framework of this paper.
Fig 2.
Schematic of RNA sequencing principle.
Fig 3.
Mean-variance relationship of real scRNA-seq data.
Table 2.
Comparison of the three data distributions.
Table 3.
The simulation data generation scheme for different genes.
Fig 4.
Mean-variance relationship of simulation dataset.
Fig 5.
Full flow of scRNA-seq data analysis.
Table 4.
Confusion matrix of different methods.
Fig 6.
Positive evaluation indicators of each DEA method on simulated datasets under different biological replications (K).
(a) K = 2. (b) K = 3. (c) K = 4. (d) K = 5.
Fig 7.
Negative evaluation indicators of each DEA method on simulated datasets under different biological replications (K).
(a) K = 2. (b) K = 3. (c) K = 4. (d) K = 5.
Fig 8.
ROC curves of each DEA method on simulated datasets under different biological replications (K).
(a) K = 2. (b) K = 3. (c) K = 4. (d) K = 5.
Table 5.
Comprehensive score of each DEA methods under different biological replications (K).
Fig 9.
Spectral clustering results of PBMCs.
Fig 10.
DE genes detected by different DEA algorithms.
(a) Venn diagram. (b) Circos plot.
Table 6.
Evaluation of different DEA methods on two real scRNA-seq datasets.