Skip to main content

Table 3 The NMI performance of cell-types prediction on different types of imputation data in three datasets. The PPI-based, pathway-based and fused imputation data were used in comparison

From: A flexible network-based imputing-and-fusing approach towards the identification of cell types from single-cell RNA-seq data

Dataset

Camp

Darmanis

Li

 

PPI

Pathway

Fusion

PPI

Pathway

Fusion

PPI

Pathway

Fusion

Raw

0.496

0.532

0.642

0.811

0.853

0.771

0.715

0.806

0.850

SAVER

0.530

0.558

0.643

0.869

0.862

0.847

0.723

0.823

0.840

scImpute

0.486

0.447

0.565

0.739

0.759

0.791

0.865

0.862

0.863

RESCUE

0.548

0.485

0.588

0.827

0.819

0.850

0.893

0.805

0.919

NetImpute

0.704

0.701

0.664

0.863

0.871

0.872

0.900

0.884

0.909

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us